Starting Out: Alekhine Defence [1 ed.] 1857443705, 9781857443707

Citation preview

I

:

alekhine's defence JOHN COX

First published in 2004 by Gloucester Publishers pic, (fonnerly Everyman Publishers

pic), :North burgh House, 10 Northburgh Street, London, EC1 V OAT Copyright© 2004 John Cox

The right of John Cox to be identitied as the author of this work has been asserted in accordance with the Copyrights, Designs and P ate n ts Act 1988.

All rights reserved. No part of this publication mar be reproduced, stored in

a

retrieval

system or transmitted in any form or by any means, electronic, electrostatic, magnetic

tape, photocopying, recording or otherwise, without prior permission of the publisher. British Library Cataloguing-in-Publication Data J\ catalogue record for thi s book is available from the British Library. ISBJ',;: 978 1 85744 370 7 Distributed in North America by The Globe Pequot Press, P.O Box 480,

246 Goose Lane, Guilford, CT 06437-0480. All other sale s enyuiries should be directed to Gloucester Publishers plc, Northburgh House, 10 t-.:orthburgh Street, T .ondon, EC 1 V OAT tel: 020 7253 7887 fax: 020 7490 3708 email: info @everymanchess.com \vebsite: www.everymanchess.com

Everyman is the registcn:d trade mark of Random House Inc. and is used in this work under license from Random House Inc.

EVERYMAN CHESS SERIES Chief Advisor: Garry Kasparov Commissioning Editor: Byron Jacobs

Typeset and edited by Firs t Rank Publishin g, Brighton. Cover Design by Horatio J\Ionteverdc. Printed and bound

in

the UK

Contents

Bibliography and Acknowledgements

4

Introduction

5

1

Main Line with 4 ... dxe5

10

2

Main Line with 4 ... Bg4

48

3

Main Line with 4 ... g6/4 ... Nc6

77

4

Exchange Variation with 5 ... cxd6

95

5

Exchange Variation with 5 ... exd6

114

6

The Four Pawns Attack

125

7

The Chase Variation

151

8

Other Tries for White

163

Index of Variations

188

Index of Complete Games

191

Bib I iography Books Alekhine's Defence,

Eales and Wi l l iams (Ba tsford 1973)

A classic a nd very usefu l for a positional understandi ng of the older systems. The Complete Alekhine

a n d New Ideas in the Alekhine Defence, both Graham Burgess

(Batsford 1 992 and 1 996)

Extremely comprehensive; the firs t in particular is indispensable. Alehhine's

Defence, Davies (Everyma n C hess 2001)

I nspi ri ng rather than

comprehensive

The Alekhine for the Tournament Player,

b u t very good on

the fashionable

li nes.

Alburt a n d Schiller (Ba ts ford 1 985)

Alburt gave little away but is interesting on his own games. Alekhine's Defence,

Hort (A & C Black 1 98 1 )

Comprehensive b u t languageless a n d a littl e du l l . However Hort is, o f course, a very strong player and his perspective i s interesting Zashchita Alekhina,

Bagirov (Fitzculture Sport 1 987)

Terrific if you have Russian; Bagirov was the foremos t Alekhine player of his day, he held nothin g back a nd his thoughts are invaluable.

Magazines and Databases Megabase 2003; www.chesspublishing.com; Chess Informant 1-87; 1WIC

Acknowledgements Richa rd Pall iser and Natasha Regan were extremel y helpful during the writing of this book, whi le Jonatha n Rogers. Peter Taylor, Wil liam Watson and D a vid Coleman also made suggestio ns. My tha nks to the m .

4

In t r o d u c t i o n

I ntroduction

Introduction to the Alekhine

Diagram 1

The Alekhine

Pl ayers see the l ig h t a n d t a k e u p t h e Alekhine for various reasons. Legend h as it th at Fischer's debut with th e defence in the 1 96 5 H a ­ v a n a tournament, i n which h e was playing by telephone, occurred when Ciocaltea shocked Bobby after 1 'd4' Nf6 with 2 e 5 . 'I thought you said 'd4', snarled Bobby. 'No, no', said the quivering monitor, ' E E E four!' Th e crowd must have expected all hell to break loose, but Fischer, i n one of those unpredictable flashes of equanimity so typical of him, shrugged his shoulders and calmly continued with 2 Nd5, the start of a productive occa sional rel ationship with the defence. Miles also took up the opening for the Capabla nca m emorial in Havana, but his explana tion was simpler: h e thought i t would be fu nny to w in Ca­ pabl anca's tou rnament with Alekhine's Defence (he did, by the way). The great Russian 'Alekhin ist', Vladimir Bagirov, was more mystical: .. .

5

Starti n g O ut: A l e kh i n e's D e f e n c e

he l iked to claim that Alekhine him self had appeared to him in a dream a n d told him to defend his opening's honour. Others have dealt better than me with the question of why to play the Alekhi n e . Burgess says t h a t it's more fun t h a n other openings. I don't know about that, but it does seem to attract the devotee. It is very no ­ ticeable in many variations which might have scored quite poorly overall how those who persevere have achieved superb results- Ken­ gis in his eponymous system, a n d Bagirov i n the 4 . Bg4 main line, for e x a m p le a n d the same is true of the opening as a whol e . It is for some reason an opening which appeals to the rebellious a nd those of a n unconventional cast of mind. . .

-

D avies poin ts out that L.Nf6 is the second best scorer in the data­ bases behind L.c5 (and no doubt it would score even higher if Kas­ parov had been playing it these last 20 years ! ). I m ight add that it is the only reply to 1 e4 with a plus score a t world championship level (trivia buffs may l ike to name the four occasions on which it has been played; answer below). Certainly it has considerable practical merit; if nothing else many White p l ay e rs are scared of it a n d duck the m a i n lines. I t also allows Black to choose the type of game to a considerable extent. There is

som ething in the Alekhine to suit all tastes; there are sharp a nd highly theoretical tactical lines, a n d it is a lso possible in a lm ost every line to be out of theory by move 6 with a solid position. With this in mind I have provided below a non-theoretical. a solid, and a tacti­ cal/theoretical re p e r toi re for Black, as well as similar suggestions for White. You will find in the following pages that I have made no great effort to attribute suggestions or analysis- informality rather than passing the work of others off as my own is the object. I have, however, taken care to attribute m y own theoretical offerings; m y motive is not vanity so much as warning the reader where he might be playing a n un­ tested move based only on my own reco m m e ndation. From time to time I refer to games I don't have space to give- this is intended to help those with access to databases, although I recognize with regret that will irritate others . The Alekhine i s very fluid as regards pawn structure and I have not given a general positional introduction for that reason. I hope general positional considerations are covered in each chapter and in the va ri­ ous Tips. The four Alekhine world championship games were Alekhine-Euwe, 2 9th game, 1 935, Spassky-Fischer, 1 3th and 1 9th games, 1972 and Ponomariov-Ivanchuk, 6th game, 2002. Al l were drawn except the first Spassky-Fischer game.

Possible Opening Repertoires Here are some repertoires worth investigating:

6

In t rodu c t i o n

Suggested Non-th eoretica l Repertoire for Black Main Line

4 . .. dxe5 5 Nxe5 c6. Presently offering a rare treat to grandmaster a n d palooka al ike. Y o u can p l a y six m o v e s o f theory a n d t h e n be m o r e or less confident that b o t h of you w i l l know no further, while B la c k is of­ fered a very sound position . Exchange variation

5 .. . exd6. Solid a n d dependable. Four Pawns

6 . . . c5 7 d5 g6. A terrific scorer for Black, with very little theory. Chase Variation

5 ... c6 and 6 ... e6. Then j ust go . . . b7 - b6 and play. Rare Fourth Moves

Glance at the main lines I give. Others

Meet 3 Nc3 with 3 . .. Nxc3 a n d 4 . . . d5 a n d meet 2 Nc3 with 2 . . . e 5 . This is if you're happy with a Vienna; otherwise there is 2 . . . d 5 3 e5 N fd 7 4 e6 fxe6 5 d4 c5 fol lowed by developing all your pieces a n d aiming at d4 a n d e 5.

Hal f a n hour looking through t h e lines and you're good to go!

Suggested Pos itiona l Repertoire for Bl a ck Ma in Line

Play the Kengis or Miles systems, or the Old M a i n Line with 4 . . . Bg4 and 5 ... e6. Exchange Variation

Play either . . . exd6 or . cxd6 to tas te. ..

Four Pawns

Learn the lines with 9 . Bg4. ..

Chase

Either a s above or with 5 . . . e6 and 6 . . . b6, a s i n Game 37. Ra re Fourth Moves/3 Nc3/2 Nc3

As above.

S u ggested Ta cti ca l Reperto ire for Bl a ck Main Line

Play 4 . . g6. Then play Alburt's exchange sacrifice after 7 a 4, or . .. Na5 lines after 7 Qe2. .

7

Sta r t i n g O u t: A l e k h i n e's Defe n ce

Exchange

P l a y the li nes with .. . cxd6. Four Pawns

Choose either 9 Be7 or 6 . . . c 5 7 d 5 e6. . . .

Chase

Play 5 . . . c6 and 6. . .d 6, a s in Game 35. Rare fourth moves

As above. Others

Meet 3 Nc3 with 3 . . . Nxc3 4 dxc3 d5, or 4 bxc3 d6. Meet 2 2 ... d5 3 e 5 Ne4 4 Nce2 f6, but th is needs work.

Nc3

with

Solid Repertoire for White Play the ma i n line and then consider the following options. Mile s System

6 Bd3, as i n Game 5, or else Be2, 0-0, Rel and so on, delaying c 2 -c4. Kengis System

6 Bc4 a n d then follow Tzermiadianos' plan i n th e notes to Game 2 . 4 ... dxe5/5 ... N d 7

Play 5 Nxe5 N d 7 and now 6 Nf3. 4 . .. Bg4 5 Be2 e6

Play the Main Line without h3 a s seen in Games 10 and 1 1.

4 Bg4 5 Be2 c6 ...

Play 6 0-0 Bxf3 7 Bxf3 a n d then continue a s in Game 1 4 . 4 ... g6

Play the line 5 Bc4 Nb6 6 Bb3 Bg7 7 Qe2. 4 ... Nc6

Play 5 c4 Nb6 6 exd6 exd6 7 h3.

Tactical/Theoretical Repertoire fo r Wh ite Play the main line and then consider the following options. Miles System

6 Be2 Bf5 a n d now go for it with 7 g4!? Kengis System

If you c a n make it work give 6 Qf3 a go, otherwise 6 c 4 . 5 ... Nd7

Get on with it with 6 Nxf7 and then choose th e 8 g3 lines. 4... B g4 5 Be2 e 6 See Bologa n's idea in Game 1 2 .

8

I nt r o d u c t i o n

4 ... Bg4 5 Be2 c6 See 6 N g5 a s i n Game 1 6. 4 . . . g6

7 a 4 and continue like Grischuk i n Game 1 8. 4 Nc6 ...

Play the l i n es with 5 c4 Nb6 6 e 6.

Offbe at Tri es fo r Wh ite 4 f4, a s i n Cha pter 8, is underrated, or 2 Nc3 w i t h Hector's ideas i s excel lent if you don't m i n d a Vienna. Whatever you choose, good luck i n your adventures with the Alekhine.

Joh n Cox, Lo ndon, October 2004

9

Chapter One

Main Line with 4

• • • • •

...

dxe5

Introduction Black Plays 5

...

g6 (Kengis)

Black Plays 5 c6 (Miles) ...

Black Plays 5

...

Nd7 (Provocative)

Illustrative Games

M a i n Lin e with 4

..

. dxe5

Introduction 1 e4 Nf6 2 e 5 Nd5 3 d4 d6 4 Nf3 (Diagram 1)

Diag r a m 1

Diagram 2

The Grandmaster choice

Black takes i mmediate action

This is White's most solid system a n d the most popular choice of When a variation has been dubbed both the Classical and the Modern, a s this one has, you can b e pretty sure that it con­ tains some timeless merit. Rather than commit himself either to maintaining the pawn o n e5 (4 c4 Nb6 5 f4) or to its exchange (4 c4 Nb6 5 exd6) White keeps his options open u ntil he sees how Black plays. Black can now either exchange on e5, the modern method, or set about development with 4 . .. Bg4 (Chapter 2), 4 . g6 o r 4 ... Nc6 (both Chapter 3) . grandmasters.

.

.

1 e4 Nf6 2 e5 N d 5 3 d 4 d6 4 Nf3 dxe5 (Diagram 2)

The revival of Black's fortunes against the main line has been associ· ated very much with this move i n recent years. It looks as though Black loses time by inviting the white knight to a strong outpost i n t h e centre of t h e board b u t , in fact, the r e s u l t can be a g a i n of time. Black's idea is that when he chall enge s the knight with . . . Nd7 White will not want to exchange a n d will have to retreat with Nf3. Thus i n effect it will b e B l a ck who h a s gained t h e move . . . N d 7 . I n addition, b y making t h e exchange before fianchettoing the king's bishop, B lack re· tains the option of changing course should White retake with the d­ pawn. Usually Black does not w a nt to fianchetto the king's bishop while the pawn o n e 5 remains because i f the pawn is solidly defended the bishop will be shut out of play. 5 Nxe5

In the event of 5 dxe5 Black's best is 5 ... Bg4, which leads to the sys· tern described in Game 14 without Black having had to play either . .. c 7-c6 o r .. . Bxf3; this is considered very comfortable.

11

Start i n g O u t : A l e k h i n e's D ef e n c e

Black plays 5 ...g6

(Kengis)

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 g6 The characteristic move o f the system. Now that the di agonal is clear Black develops his king's bishop t o its best square. The price he has to pay for this i s that usu ally he will need to give up his queen's bishop s in c e once the pawn comes to g6 there is no suitable outpost available for it. The system is ri ghtly named after Latvian grandmaster Edwins Kengis. H e was not the first to play i t bu t his fine results i n the pe­ riod 1 988 - 1 99 6 b rou gh t the system general acceptance. White now h a s v a rious choices. The mai n line is 6 Bc4 (Games 1 and 2), while 6 c4 is the other ma i n try (Game 3). W h i te also has a sharp move in 6 Qf3 (covered in the notes to Game 3) and quiet approaches i n 6 Be2 a n d 6 g3 (see the notes to Game 1). White plays one of the positional l i nes i n Games 1 a n d 2, the Kengis i s really quite a non-theoretical a n d simple sys tem. The probl em is learning to play the resulting positions well. Kengis was a master of this a n d playing through all his games in the system on a database is time well spent. The typical position is that which arises after, say, move 17 in Game 2 (Diagram 1 5) . There a re variations i n t h e placing o f the a - p a w n s a n d o f course of t h e pieces b u t this type of position, with Black's queen's bishop exchanged for a knight, will arise more often than not. Black is rock solid and well placed to meet any aggressive attempts, but what are his posit ive ideas? Assuming that

This open ing system as a whole reminds me of Michael Stean. Mi­ chael's an accountant now, I believe- millionai re, no dou bt- but when I was a lad he was a grandmaster whose lucid classic Simple Chess influenced a generation of English j uniors. The Kengis system didn't exist i n those days but if it had Stean would have instantly rec­ ognised it as a perfect exemplar of two theories he expounded i n that book. The first is that the structure where Black has exchanged his 'd­ and c-pawns for Whi te's 'd-' and e -pawns is i n some sense fundamen­ tally favourable for Black. Stean's point was that Black can advance his pawns o n the queen's wing - a minority attack, in fact- and hope to exchange down to a single weak white pawn there, and meanwhile Black h a s a n extra central pawn which gives him a better chance of establishing a protected outpost in the centre, or alternatively he can advance the central majority and use the 'e-' and f-pawns to expose White's king . This idea - the anti thes is of the class ical view that a queenside pawn majority was i n itself a good thing- was something of a revelation to a schoolboy player brought up in the 1 970's on Euwe and Pachman. As a hard a n d fast r u le it's nonsense, of course (as Stean would be the first to say), but it's a useful insight as a way of thinking about the game, and i n particular about this opening. Black wants to play .. . c7-c5: this is his fun damenta l break ( ... e7-e5 may equal ise completely but is not usually playable). He will then have

12

M a i n lin e w i t h 4

...

dxe5

exactly the structure Stean was talki ng about, and most often he will already have advanced his a-pawn a nd thus have made some progress with a minority attack. He will also h a ve the knight on d5. This i s dif­ ficult to drive away by c2 -c4 without White creating weak nesses. For example if White has met the a-pawn's o nslaught with a 2-a4 then c2c 4 will allow the k night i nto b4 a n d s i multaneously cripple White's majority, while if he has allowed Black's a - p aw n to a4 before playi ng a2-a3, then c2-c4, while it will evict the knight, will leave the c-pawn exposed without pawn protection, a n d t he b-pawn backward. These ideas may seem rather abstract, but they can frequently be ob­ served in endgames or late middlegames arising from this opening, and so can the principle that Black must use his own pawn majority posi­ tively- once White's d-pawn has been removed . . f7-f6 and . . . e7-e5 is a common theme (see Mukhametov-Bagirov, Bern 1 995, for example). .

Stean's second theory was that some, spatially inferior, pawn struc­ tures have capacity for a certain number of pieces, and that if the de­ fender can exchange down to that number the space adva n tage won't matter. Black's pawn structure in the Kengis is a classic example of this- Black has space for three mi nor pieces to be deployed effi­ ciently, but not four, hence the fate which invariably awaits the ' queen s bishop in this system. Indeed this is something o f a theme of the Alekhine as a whole: perhaps I might call it the 'three piece rule.' It may sound simple, but it w a s realising this which e nabled p l ayers to see that 4 . . . dxe5 5 Nxe5 gained time not for White but for Black.

TIP: I n this system Black must try to exchange a minor piece: i f pos­ sible a knight for the one on e5, or the light-squared bishop for the one on b3, but bishop for knight will do.

Black Plays 5 c6 (Miles) ...

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 c6 (Diagram 3 )

This i s the hottest variation of the Alekhine right no w . The la te Tony Miles w a s not the first to play it, but his successes with it from the mid-nineties have made it the trendiest way to play, ahead even of the Kengis system . Like the Kengis, the most obvious motive is to prepare .. . Nd7 without allowing Nxf7 (see Game 8). Compared to the Kengis, the merit of 5 ... c6 is that Black preserves the options of his king's bishop, a nd hence also of his queen's bishop - in the Kengis this piece almost always h a s t o be exchanged, s i n ce once i t develops to g4 and i s challenged with h3 i t ca nnot retreat to h 5 i n view o f g2-g4. 5...c6 also makes t he knight on d5 more flexible: one day White will play c 2-c4 and, when he does, the additio nal option o f retreating to c 7 w i l l be available, a post from where it is less easy to shut the knight out of the game than is the case with b6. The drawback is perhaps that . .. c7- c6 loses a little time- in t he Kengis ... c7-c6 is played only out of necessity i n reply to Bc4.

13

S t art i n g O ut: Ale k h i n e 's Defe n ce

Diagram 3

Diag r a m 4

The latest trend

Black invites Nxf7

White has a wide c h oic e of moves after 5 . . . c6. H e can develop his king's bishop to e 2 d3, c4 or even g2, he can drive the knight away ,

immediately with c 2-c4, or he can play N d2 , intending to bolster t he e5-knight with N2 f3. O f these, 6 Be2 (Game 4) is Kasparov's choice a nd by s ome dista nce the main line.

Black Plays 5 Nd7 (Provocative) ...

1 e4 N f6 2 e5 N d 5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 Nd7 (Diag ram 4 )

5 ... N d 7 i s a move w i t h a n entertaining history. I t first appeared at a high level i n the 1 96 5 Candidates Match between Tal and Larsen . O n seeing t h e knight arrive o n d7 T a l w a s unable t o believe h i s eyes a t t h e opportu nity h e w a s bein g offered, thought for 4 5 m inutes, and continued q u ietly instead of sacrificing ... i n the end he barely m a n· aged to draw. Subsequent a nalysis purported to prove that 6 Nxf7 Kxf7 7 Qh5+ Ke6 was winning for White (see Game 8), a n d the variation disappeared until the early 1 980's, when people started to play it as Black, and to

w i n . Today the theory o n the sacrifice is extensive: to be up to date you need a good knowledge of Polish correspondence chess, a n d no doubt somewhere there are websites devoted to it. At the moment, however, theory favours Black, and the line which is giving Black the most trouble is actually the patient 6 Nf3, which w e see in Game 9. The variation does have two significant practical drawbacks: first, White can d raw at o nce with 8 Qg4+ Kf7 (8 . .. Kd6 is suicide) 9 Qh5+; secondly, you have to know a fair amount of theory which you will rarely g e t t o p l ay, since White is not likely to venture the sacrifice off the c u ff. On the other hand, it is quite possible that Black's fou rth a nd fifth moves are objectively the best: it is hard to believe that Black should be much worse, if at all, after 5 ... Nd7 6 Nf3.

14

M a i n L i n e w i th 4 . . . d x e 5

Illustrative Games Game 1 D L .Christ iansen • E . Kengis Manila 1 992

1 e 4 Nf6 2 e 5 N d 5 3 d4 d6 4 Nf3 dxe 5 5 N xe5 g6 6 Bc4 (Diagram 5)

Diagram 5

Diagram 6

Active development

B lack prepares ... Nc7

We should note that 6 g3 is perfectly sensible, after which Black should steer for ... e7-e5 rath er than push the c-pawn. For example 6 ... Bg7 7 Bg2 0-0 8 0 -0 c6 9 Re1 Bf5 1 0 c3 Nd7 1 1 Nf3 Re8 1 2 Nbd2 e 5 ! 1 3 N c 4 Bg4 is level. Altho u gh ...e7-e5 i s rare i n the Ken gis White's bishop on g2 does not restrain it by pressure on f7 as the b 3 bishop does in the main line, and it is also useful to keep the c6-b7 pawn bar­ rier intact against White's b ishop. 6 Be2 is not a theoretical option but has been play e d by some good pla y er s . Then 6 ... Bg7 7 0 - 0 0 - 0 8 c4 Nb6 is na tural, a n d now BJack

has reasonable play after both 9 h3 Nc6! and 9 Nc3 Be6 10 Bf3 ( 10 b3 c 5 is t h e note to Wh ite's n inth in Game 3) 10 ... c6 1 1 b3 f6! 12 N g4 Bxg4 1 3 Bxg4 f5, winning the d 4 - pawn. 6 ... c6

D educt poi nts for 6 . . . Bg7?? 7 Nxf7 Kxf7 8 Qf3+. WARNING: In these ...dxe51ines always re member to watch out for

Qf3, especial ly when the d5-knight is un protected.

7 0-0 B g 7 8 Rei

Wh ite's a lternative 8 Nd2 is seen in the next game. 8... 0-0

9 Bb3

This i s Wh ite's most p opu la r choice a t this point, a lthough he c a n a lso

15

S t a r t i n g O u t: A l e k h i n e ' s D e fe n c e

play 9 c3. Black contin ues in the same way with 9 . . . Be6 10 Nd2 (for 10 Bb3 see the next note) 10 . . . N d7 . Now 1 1 Nd3 N c 7 forces the even more desirable trade of bishops, while 11 Nef3 Bg4 leads to very typi­ cal positions s i mi lar to the next game. 1 1 Nxd7 Qxd7 is ha r m le s s al­ though Black's play i n van der Wiel- Bagirov, Erev a n 1996, is worth remembering: 12 Ne4 b6 13 Ng5 Bf5! 1 4 h 3 f6! 15 Nf3 Rae8. Black then took h is bishop to f7, with a very solid position in w h ich he can slowly arrange for ... c6-c 5 . 9 Be6 (Diagram 6 ) 1 0 N d 2 •..

1 0 c 3 N d 7 1 1 Nf3 N c 7 1 2 Bxe6 (after 1 2 B c 2 Black can either play 12 . . . c5 or 12 . . . Bd5, a fter which h e may even b e able to aim for . .. e7-e 5 ) 1 2. Nxe6 is much t h e same ( 1 3 Q b 3 Qb6). A purposefu l move i s 1 0 h 3, aiming to play 1 0 . . . Nd7 1 1 Nf3 and a void the . . . Bg4 pin. Black contin­ ues as usual with 1 l...Nc7; if he is allowed to exchange the light­ squared bishops he is doing fine, while after 1 2 c 4 he can continue with Davies' 1 2 . . . Nb6 1 3 Qe2 a 5 1 4 a 4 Nc8!, intending either . . . Nd6-f5 or . . . b 7 -b 5 . Th is position well ill ustrates some themes of the system: once White is provoked into c2-c4 his d-pawn is weak and his majority on th e queenside has been crippled; in additio n the . . . b 7-b 5 strike, trading away the c-pawn, isolating the d-pawn and securing the mighty d 5 -square, is typical. ..

1 0 N d 7 1 1 Nef3 Nc7 ..•

Black cou ld a l s o go 1 l . . . Bg4 with the usual sort of game, but i n this exact position h e can get . . . c6-c5 in a n d trade the light-squared bish­ ops.

Diagram 7

Diagram 8

Time for a pawn break

A critical position

1 2 c3

1 2 N c 4 a 5 1 3 a 4 Bd5 is also fine. 1 2 ... c5 13 Bxe6 Nxe6 14 d5 Nc7 1 5 Ne4 Nf6 1 6 c4

Mter 1 6 d 6 exd6 1 7 Nxd6 Ncd5 1 8 Nxb7 Qc7 1 9 c 4 Qxb7 20 cxd5

16

M a i n Lin e w i t h 4

...

dxe5

N x d 5 Black's active pieces easily compensate for h i s slight structural disadvantage. 16 N xc5 Ncxd5 17 Nxb7? Qb6 is the same theme. 16 ... Nxe4 17 Rxe4 ( D iagram 7) 17 ... e6

1 7 ... b5 would b e the other thematic break, b u t here t h e equally typical reply 1 8 Bg5 would b e strong. 1 8 Bg5

18 d6 Ne8 19 B f4 Qb6 leads White nowhere. 1 8 ...f6 1 9 d 6 Ne8 (D iagram 8) 2 0 Rxe6

critical moment. 20 d7 fxg5 2 1 dxe8Q Rxe8 favours Black, but could h ave baled out with 20 Be3 Qxd6 2 1 Qxd6 Nxd6 22 Rxe6 Nxc4 23 Bxc5 with a d rawn endgam e . The text looks like a rash ventu re, b ut Ch risti ansen does okay with it in th e game.

A

Christiansen

20 ... fxg5 2 1 Qd5 Kh 8 22 Rae 1

2 2 Nxg5 is met by 22

. . .

Bf6, freeing

the g7 -square for the

king.

2 2 ... Nf6 2 3 Qxb 7 Q b 6 2 4 R e 7 Qxb7 2 5 Rxb7 Rfe8 2 6 R x e8+ Rxe8 (Diagram 9)

Diagram 9

Diagram 1 0

Decision time

Black should be winn ing

27 h 3

could - a n d surely should - have baled out with 27 d7, when 27 . . . N x d 7 is forced a s 27 . . . Rd8? 28 Ne5 Kg8 29 Rxa7 l eaves Black totally tied up, b u t 28 Rxd7 g4 29 Nd2 gives Black a t least ade­ quate play for his missing pawn, for example 29 . . . Bxb2 30 Rxa7 Re2 3 1 Nfl Bd4 32 N e3 R e 1 + 33 N fl Re2 w ith a repetition.

Again Wh ite

2 7 ... Ne4 28 d 7 R d 8 2 9 Rxa7 Bxb2 30 a4 Nf6 3 1 Nxg5 Rxd7 3 2 R x d 7 Nxd7 (Diagram 10)

Black must be winning th is, but Ch ristia nsen puts up a good fight. 33 a5 NbS 34 Ne4 N a 6 35 f4 Ba3

I t might have been better to b ring the king u p rather than go imme-

17

Starti ng O u t: Ale k h i n e ' s D e fe n ce

diately for t h e a - paw n, but only analysis of some complicated pawns versus pieces positions in twenty moves' time could really tell us. 36 Nd6 Bb4 37 N b 7 Kg7 38 Kf2 Kf6 3 9 Kf3 Ke7 40 Ke4 Kd7 41 Kd 5 Nc7+ 42 Ke4 Na6 4 3 Kd 5 Kc7 44 Nd6 Bxa 5 45 Ne8+ Kd8 46 Nf6 N c 7 + 47 Kd6 Bd2 4 8 g3

Not 48 Nxh7 Bxf4+ 4 9 Kxc5 Ke7 5 0 g4 Ne6+ when White's knight is sidelined a nd Black wins. 4 8 ... h5 49 Ne4 Be3 50 h 4

Again, n o t 5 0 Nxc5 h 4 5 1 gxh4 Bxf4+ when Black's pieces combine well to drive White's king from its present dominating position. 50 ... Bd4 51 Ng5 Bf'2 52 Ne4 Bel 53 Ke5

After 53 Kxc5? Ke7 54 Kb6 NeB 55 c5 Ke6 56 c6 White wins the k night, but Black's king marches into the kingside and he wins. 53... Ke7 54 f5 gxf5 55 Kxf5 ?

White could have dra w n with 5 5 Nxc5! Bxg3+ 56 Kxf5 Bxh4 5 7 Kg6 Bf2 5 8 Nd3 ! here. It is difficult t o believe that Black could n't have won this ending at some point. Kengis i ndicated 45 ... Kb6 a n d 4 7 . . . Ne8+ a s earlier tries. 5 5 ...Na6!

Not 5 5 . .. Ne6 a s Black needs this square for his king. 5 6 Kg6 Ke6 57 Ng5+

5 7 Kxh5? Ke5 is easy for Black. Instead C hristia nsen drives back the king because of the trick 57 Ke5? 58 Nf3+, and continues his ingen­ ious resistance. . . .

57 ... Ke7 58 Ne4 Ke6 5 9 Ng5+ Kd7 60 Ne4 Kc6 61 Kxh5 Nc7 62 Kg6 N e 8 6 3 h 5 Nd6 64 Nxd6

White has done everything he ca n bu t it's still lost. 64 Nf6 N x c 4 65 h6 Ne5+ 66 Kf5 (66 Kg7 Bc3! 67 h7 Nd7 and . . . Bxf6+) Nf7 67 Kg6 Nx h6! 68 Kxh6 c4 6 9 Ng4 c3 70 Ne3 Bd2 is a particularly cruel point. 64... Kxd6 6 5 g4 Ke6 66 Kg7

Nothing help s: 66 g5 Bd2 67 h6 Be3 68 Kh5 Kf5 69 g6 Kf6 7 0 g7 Kf7 illustrates the winning method. 66... Bc3+ 67 Kf8

Or 67 Kg6 Ke5 with a s i milar fi nish to the last note. 67 ... Kf6 6 8 Ke8 Bd2 69 Kd7 Ke5 7 0 Ke7 Be3 7 1 Kd7

7 1 Kf7 sets an i nstructive trap: 7 L..Kf4 72 h6 Kxg4?? allows a draw wit h the switchback 73 h7 Bd4 7 4 Ke6 Kg5 75 Kd5 Kg6 76 h8Q, but the cool 72 . . . Kg5! wins . 7 1.. . Kd4 7 2 Ke6 Kxc4 73 Kf6 Kd5 74 g5 c4 75 g 6 c3 0- 1

It's close but Black wins: 76 g7 Bd4 + 77 Kf7 Bxg7 78 Kxg7 c2 79 h6 c l Q 8 0 h 7 Qc7+ 8 1 Kg8 Ke6! 8 2 h8Q Qf7 mate. A terrific battle. The ending was amazingly difficult to wi n for Black. Christiansen de­ fended him self su perbly wit h the exception of one lapse.

18

M a i n L i n e w i th 4 . .. d x e 5

Game 2 0 Z.Aimasi • Kengis Bern 1 996

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 g6 6 Bc4

6 N d2 a t this stage is possible: 6 .. . Bg7 7 Ndf3 0-0 8 c 4 (8 Bc4 c6 is the ga m e) Nb6 9 Be2 N8d7 10 Bf4 c5 is fine for Bl ack after either 1 1 d xc5 Nxc5 or 1 1 d 5 Nxe5 12 Bxe5 Bxe5 1 3 Nxe5 Qd6 14 N f3 e6. 6 ... c6 7 0-0 Bg7 8 Nd2 (Diagram 1 1)

Diagram 1 1

Diagram 1 2

White prefers knight play

Danger!

Wh ite intends to sen d this knight to f3. Sometimes he will exchange th e other knight, but it is usually thought that Wh ite does better to retain more pieces by retreating to d3, as Almasi does. In fact White's move order is not accurate. With 8 . . . Nd7 at this poin t Bl ack can force an exchange o f kn ights since 9 Nd3? l eaves the d- pawn ha nging. If White wa nts to b e sure of getting the version with Nd3 h e should play 7 Nd2 Bg7 8 Ndf3. 8 . . . 0-0

9 Ndf3 N d 7 10 Nd3

10 Re1 Nxe5 1 1 Nxe5 i s the a lternative. White can then arrange his pieces comfortably w it h Qf3, Bg5 and Rad1, but Black can arrange his conveniently also , for example with 1l...Be6 (D iagram 1 2). Now 1 2 c3? Bxe5 a n d 1 3 . . . Nxc3 a t least equalizes for Bl ack. WARNING: When Black plays ... Be6 with his knight on d5 and White's bishop on c4, watch out for the trick with ...Bxe5 and ...Nxc31

1 l . . . Be 6 1 2 Bb 3 a 5 1 3 a 4 Qc8 1 4 c 3 Rd8 h a s been played in many games, with equality. Black can continue, for instance, with .. . b7-b6 a n d . . . Ra7 to lend sup port to e7, a n d then arrange ... c6-c5. 1 l . . . Be6 1 2 Bfl a 5 1 3 a 4 Qc8 1 4 c 3 Rd8 i s much the same. 10 Re 1 Nxe5 1 1 dxe5 is also possible, of course, but a llow ing the open

19

S t a rt i n g O u t: A l e k h i n e's D e fe n c e

d - file i n this fashion rarely ach ieves anything i n t h e Kengis Syste m, a n d this posi tion i s n o exception: l l. . .Bg4 1 2 h 3 Bx f3 1 3 Q x f3 e 6 is al­ ready equal, Black continuing with . . . Qc7 and doubling on the d-fil e. 10 a5 ...

(Diagram 13)

Diagram 1 3

Diagram 1 4

Black's logical advance

Black's problem bishop

A thematic mov e . Black fo rces White to weaken his queenside to pre­ vent the threat to trap the bishop, and sets in motion the usual mi­ nority a ttack. 1 1 a4

One drawback of this move i s that it is now n ever goin g to be possible to push Black's knight away with c2-c4 without giving i t the juicy b4square. For this reason 1 1 a 3 is sometimes preferred. This was pl ayed in Tzermiadianos-Kengis, Komotin i 1 992, in which Kengis' unbeaten streak in the o pe n in g came to a n end. Whi te's plan i n th a t game w a s s i m pl e and direct: c2 -c3, Qxf3, Re l , Bd2, Radl , Be l, Bbl, Nf4 a n d then launch t h e h - p a w n and try to d eliver mate. Kengis failed t o find his u s u al counterplay and went down in flames. This is a simple and dangerous way for White to play; I have space only for the moves b u t the g a m e is worthy of investigation for prospective White players: 1 1 a 3 N 7 b 6 1 2 B a 2 Bg4 1 3 c3 e6 1 4 h 3 Bxf3 1 5 Qxf3 N c 8 1 6 Re l Nd6 1 7 Bd2 Nf5 1 8 R a d l Qc7 1 9 Bel b6 20 Nf4 Rfd8 2 1 Bbl Nd6 22 h4 b 5 (22 . . . h 6!?) 2 3 h5 b 4 2 4 hxg6 h xg6 2 5 Qg3 N x f4 26 Bxf4 bxa3 2 7 bxa3 Qd7 28 Be5 Ne8 29 Be4 a4 30 Rd3 N d6 31 Qh4 f5 32 Bxc6 Qxc6 3 3 Bxg7 Kxg7 3 4 Qe7+ N f7 3 5 Rxe6 R d 6 36 Qf6+ Kg8 3 7 Qxg6+ Kf8 3 8 d5 1 -0. Black n eeds to prepare . . . c6-c5 rather more directly th a n Kengis did i n tha t game, perh a ps with th e same . . . Nd7 plan a s in the Almasi game; his . . . b 5-b4 was not as effective since by the time it arrived White could ignore it. It is harder to ignore . . . c6 -c5 since losing the d­ pawn is a lot more serious for White than losing, for example, the a ­ pawn. Nor w e re th e wanderings o f the b 6 knight helpful: f5 is only a

20

M a i n L i n e w i th 4

..

. d xe 5

good sq uare for this if i t either inconveniences a bishop on e 3 o r a t­ ta cks the d-pawn a fter c2 -c4 . ll

...

N7b6 1 2 Bb3 Bg4 ( D iagram 14)

There is a difference of opinion about whether Bl ack should play i n ­ stead . .. Bf5 xd3, elimi n ating th e k n i g h t w h i c h is holdi ng b ack . . . c6-c5. Kengis switched to the text after trying . Bf5, a n d for that reason I'd go with it too, b u t there's prob ably not much to choose between the two. ..

1 3 c3 e6 14 h3 Bxf3 1 5 Qx£'3 Nd7

characteristic regrouping. Th e knight was doing noth ing on b6. Sometimes it goes via c8 t o f5, bu t from d7 it supports c5 a nd a l so h as th e option of going to f6 to h e l p defend on the kingsi de. Another

16 Rel Q b 6 1 7 Bc4 h6 (Diagram 1 5)

Diagram 1 5

A

useful

nudge of the h-pawn

Diagram 1 6

White toys with

h4-h5

This position had been reached before and this move was Kengis' nov­ elty which, again, is typical of his handling of the opening. Often White will play h 3-h4 in an attempt to get somewhere on the kingside, and of­ ten Kengis meets this with . .. h7 -h6, intending to answer h4-h5 with ...g6-g5, rather than panicking and putting his own pawn on h5 as l esser players often do (this serves only to help White open lines on the kingside). Another benefit of . . . h 7-h6 is to control g5 so that a rook can come to the d -file in peace without being annoyed by Bg5 . 1 8 Bd2 Rad8 1 9 R e 2 R fe8

Kengis has decided not to prepare . . . c6-c5, instead placing his pieces where they restrain Wh ite, waiting. Bl ack is imm ensely solid in these positions; White can claim he is slightly better with the two bishops and a space advantage, but watch how far even a player of Almasi's cl ass gets. 20 Rae 1 Qc7 2 1 h 4 Qc8

21

Starti n g O u t: Ale k h i n e's Defe n ce

N o t . . . h6-h 5. 22 g3 b6

Black can't arrange ... c6-c5 due to t he pressure on d 5 and down t h e e­ file, a nd he doesn't care to p l a y .. . N 7 f6 and allow Ne5. But he can simply wait until W hite commits his forces elsewhere. Mea nwhile he m a kes useful moves. 23 Kg2 Qb7 2 4 Bb3 Qa6

Pressing an noyingly on the loose d3-knight. 25 Be l

2 5 c4 Ne7 does n't h elp White as then the d-pawn needs looking after. 2 5 ... Rf8

Unpin ning the e-pawn 26 Nf4 N 7f6 27 Nd3 N d 7 28 Rh l (D iagram 16)

I suppose the i d ea of this was at some point to play h5 and meet g6g5 with Bxg5 hxg5, h 6 etc. One can imagine tha t if White had a bishop o n c2 , for example, this might give quite a dange rous attack, ...

since ... Bf6 blocks both the f6-square and the f-pawn and might allow Qe4-h 7 +, while . . . Bh8, h 7 + Kg7, Qh5 might also be awkward. 28 ... c5! Exactly so! Black has arranged his pi ec es so that as soon as Al masi tries to get somewhere on the kingside h e has this equalizing break.

2 9 dxc5 b x c5 30 Bxd5

This i s necessary a s t h e threat of ... c5-c4 a nd ... Ne5 -d3 was too strong. 30 ... exd5 3 1 R e 7

I ' m not a t a l l sure about this. I t seems to m e i t w a s high t i m e to recall the other rook from its h 1 outpos t . Black's t hreat is . . . c4 and . . . Ne5, and to my mind 31 Rhe 1 was the natural way to prevent t hat. 3 1 .. .c4 32 N e l N e 5 ?!

This wins material and is possibly objectively the best, but I suspect Kengis had underestimated the play White now enjoys. Instead I think 3 2 . . . Qd6 was the p ractical move- the rook has to retreat a nd then 33 Re2 Nc5 se e ms to leave Black considerabl y better, White's knight on e 1 being a particularly poor spe c i m e n and his a - pawn being marked for death. 3 3 Rxe5! Bxe5 3 4 Bxh6 Rfe 8 35 Nc2 It's entirely possible Black has a good defensive seque nce here but a sudden change i n the pace of the game like this is always difficult to

cope with. 35 ... Qc6 36 Bg5 (D iagram 1 7) 3 6... Rb8

A com mittal move. I nsti nctively I would prefer 36 ... Rd6, but I im agine these moves were being made during a time scramble, and a deep

22

M a i n L i n e w i th 4 ... d xe 5

a n a lysis wou l d b e required t o say what i s b es t. 37 h5 Rxb2 38 Ne3 Qe6 3 9 Ng4 Bg7 40 h 6 f5

Reaching the time co n trol, but by now things wrong.

have

definitely gone

Diagram 17

Diagram 1 8

Where should the rook go?

The rook decides

4 1 Ne3 Bh8

This s e ems a bit feeble to me; Black still had some defensive cha nces after 4L.. Bxc3 42 h7+ Kh8 4 3 Nxc4 Qe4 4 4 Nxb2 Bxb2 - 4 5 Bd2 Qxf3 46 Kxf3 Re4 47 Rh4 Bf6! might be an important resource - and I don't quite see how White improves o n th a t. 42 Nxd5 Qe4 43 Qxe4 fxe4 44 h7+ Kf8 45 Rh4 (Diagram 1 8)

Decisive. The t h reat of 46 Rf4 + forces Black to start shedding paw ns. 45 ... Reb8 46 Rf4+ Ke8 47 Ne3 Bxc3 4 8 Rxe4+ Kd 7 49 Nxc4 Rc2 5 0 Re7+ Kc6 5 1 Bf4 Rf8 5 2 Rc7+ Kd5 53 Ne3+ 1-0

A game very characteristic both of th e system a n d of Kengis' great skill with it, as h e outplayed an opponent who has been among the top 20 in the world, a lthough in the e nd Almasi's greater strength told.

Statistics White has 63% with 6 Bc4, but i t is noticeable how well playe rs who persevere w ith the system do -like Kengis, who after four losses i n h i s ju nior days took the system u p regularly i n 1 986 a n d didn't lose a game i n my database until 1992 (over 25 games).

rh eo retical? No. The usual type of m iddlegame arises all but automatically. You need to k now w h a t to do then, though.

23

Starti n g O u t: A l e k h i n e ' s D e fe n ce

Theoretica l Statu s Solid. The weight of the struggle will always be i n the middlegame. Game 3 0 G rischuk • Baburi n

Torshavn 2000 1 e4 N f6 2 e 5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 19)

g6 6 c4 (Diagram

D iagram 19

D iagram 20

White evicts the knight

Black delays ... c7-c5

A direct but rare try. Another rare (fortunately) but very dangerous try is 6 Qf3. I n fact t his is the only move which might conceivably re­ fute t he Kengis. The idea is to grab material after 6 .. . Be6 7 c4 Nb4 8 Qxb7 (8 d5 Nc2 + is nonsense since White has to put his king on the d­ file after w hich the pin prevents dx e6- 9 K d1 Nxa1 1 0 Nd2 Bf5 1 1 Bd3 Nd7). Bl ack is supposed to have a strong reply in the exchange sacrifice 8 . . . Qxd4 (8 ... Nc2 + is very complicated) 9 Qxa8 Qxe5+ 10 Kdl N4c6, when n atural moves give him a strong attack, e.g. 1 1 Nc3 Bg 7 1 2 Be3 0-0 13 Qb7 Rd8 14 K e 1 Nd4 15 Bd3 Nbc6 16 Nd5 Nxd5 1 7 cxd5 Qxd5. This h a s o nly bee n played i n o n e game, however, a n d I don't think t he story is by a n y means over. The next page might be Fritz's 10 Be2 instead of 1 0 Kdl, for example 1 0 . . . Bxc4 1 1 Qxb8+ Kd7 12 0 - 0 Qxe2 13 Nd2, or 10 ... Nc2 + 1 1 Kd1 Nxa 1 1 2 Qxb8+ Kd7 1 3 Qxa 7 . I haven't seen t his suggested anywhere but i t looks quite inter­ esting to me . At least White isn't the only one with his kin g wander­ ing about the centre . . . 6... Nb6 7 Nc3 Bg7 8 Be3

White addresses . .. c7-c5. The other way to try and prevent this is with 8 Bf4 . Black can then continue with either 8 . . 0-0 9 Be2 c6 1 0 0-0 Be6 1 1 b3 N8d 7, with a very solid pos iti o n or Kengis' cunning 8 . 0-0 9 Be2 a5, when his idea is 1 0 0 - 0 Nc6 1 1 Nxc6 bxc6 12 Be5 Bxe5 1 3 .

,

24

. .

Main Line with 4

..•

d xe 5

dxe5 Be6 1 4 b 3 Qxd1 1 5 R a d 1 a 4 w i th a typical endga m e which i s probably about equ al . 8 0-0 (Di a gram 20) ..•

8 ... c 5 w a s alw ays Kengis' choice b u t 9 dxc5 Q x d 1 + 10 Rxd1 Bxe5 1 1 cxb6 Bxc3+ (best- 1 l . . .axb6 1 2 Bd4 is just better for White) 1 2 bxc3 axb6 le a d s to an endgame which is not too exciting for Black a n d quite possibly nastily worse, a n a tural continuation being 1 3 R d 2 N d 7 14 Be 2 Nc5 15 0-0 Bf5 1 6 Bf3 0 - 0 and now White h a s a n u m b e r o f tries, e.g. 17 Rel. It may be th a t Black h a s a defence t o a ll of these but t here must be more to life than d efendi ng this sort of position. 9 Rc1

This is a fine move. To un derstand the idea we need to look at the m ost obvious a l ternative, 9 Be2. This is rather well met by 9 . . . Be6 1 0 b 3 (10 ... Bxe5 was threatened, winning th e c-pawn, and 1 0 d 5 Bxd 5! 1 1 Nxg6 Bxc3+ 12 bxc3 hxg6 1 3 cxd5 Qxd5 wins a pa w n) 10 ... c 5 . Now 11 Rc 1 is a n inferior version of the text ( 1 l . ..cxd4 12 Bxd4 f6 13 N f3 B h 6 1 4 Ra 1 Nc6 15 Bc5 N d 7 develops sm oothly, e.g. 1 6 B a 3 Q a 5 1 7 B b 2 Ra d8), w h i l e 1 1 0 - 0 a l lows 1 1 ... Nc6! 1 2 Nxc6 bxc6 when White i s h ampered b y the undefended c3-knight, e . g . 1 3 dxc5 Bxc3 1 4 Rc 1 Bg7 1 5 cxb6 axb6 with equality. The poi n t o f 9 Rc l (Nigel Short's move) is to avoid the n eed to m eet 9 . . . Be6 with 10 b 3 - i n stead 9 Rc1 Be6?! runs i n to the e ffective 10 Ne4. 9 f4 i s sharp, aiming to maintain the e 5 -kn ight where it is a n d castle queenside. B lack shou l d still aim for ... c 7-c5, al though 9 ... c5? 10 dxc5 is obviously not the way to do this. 9 . . . N6d 7 is a good move (better than 9 . . . N8d7 since this a llows the b-pawn to b5 a n d the queen to a 5, and l eaves the b 8 -kn ight in place to com e to c6), when one game con­ tinued 10 Qf3 c 5 11 0-0-0 Qa5 1 2 Nxd7 Nxd7 1 3 Bf2. Black now h a d a choice of two good plans- 1 3 . . . Re8 14 Be2 cxd4 15 Bxd4 e 5 , com plet­ ing the d emolition of Whi te's centre and at least equalizing, and 1 3 . . . Rb8 1 4 Be2 b 5 with fierce play. In the gam e Black lost the plot and m ixed these ideas with 13 . . . Rb8 14 Be2 cxd4 15 Bxd4 e5?!. 9 Qf3 i s a nother sharp m ove, b ut Agdestein's fi n e reply 9 . . .f6 10 N d3 e5 1 1 dxe5 fxe5 12 Q d 1 Nc6 leaves B l ack already better w ith the knight coming to d4. 9 ... c5 10 Be2

White m ight still seek a modest edge with 1 0 dxc5 a n d either 1 0 . . . Qxd 1 + 1 1 K.xdl Bxe5 1 2 cxb6 a xb6 1 3 Bxb6 Bxc3 1 4 Rxc3 Rxa2 15 Kc2, or 1 1 Nxd 1 Bxe5 12 cxb6 axb6 1 3 a3, but this hasn't been tried.

1 0. . cxd4 1 1 Bxd4 B h 6 .

1 1 . . . Be6 still does n o t force 1 2 b 3 ? - instead White replies 1 2 Bf3!. Bl ack m ight have tried to invert the order of his next two moves with 1 1 . . .£6. I think the a nswer to that is not 12 Ng4 preventing ... Bh6, when 1 2 .. . Nc6 1 3 B c5 Qxdl+ 1 4 Nxd 1 ( 1 4 Rxd 1 Bxc3+ 15 bxc3 N a 4 wins a pawn) 1 4 . . .f5 1 5 Nge3 N d 7 1 6 B a 3, for example, i s reasonable

25

S t a rt i n g O u t : A le k h i n e ' s D e fe n c e

for Black, but 1 2 N f3 Bh6 w h e n , surprisingly, 1 3 Be3 B x e 3 1 4 fxe 3 i s quite strong a s White's d evelopment and ideas of Qb3, 0-0, Rd l a n d Nd5/Nd4 a r e hard t o handle. 1 2 Ra1 f6 13 Ng4! (Diagram 2 1 )

Diagram 21

Diagram 22

Uncompromising play

White maintain s the pressure

13 Nf3 Nc6 1 4 Bc5 Be6 is much l ess pointed since 15 b3 is met by 15 . . . N d 7 followed by . . . Qa5, as mentioned above. 1 3 ... Bxg4

Th e only sensible move. 14 Bxg4 Nc6 15 Bc5

Grisc huk refrains from t h e tempting 1 5 Be6+ K h 8 16 Bc5 when 16 . . . Qc7 causes difficulties with its threat to fork the bishops by .. . Qe5. White then has to play 17 Qe2 when 17 Ne5 18 b4 a 5 gives Bl ack coun terplay ( Black played his king to h8 and not g7 so that Nb5-d4-e6 in this positio n should not b e check ) . .. .

15 Ne5?! ...

This move does not work out. I think Black should have played 1 5 . . . f5 a s the bishop is just too strong once it la nds on e6. Then 1 6 Be2 ( 1 6 Bf3 Ne5 i s annoying) 1 6 . . . Qxd l + 1 7 Rx d 1 Bg7 i s interesting - White cannot defe nd the c3 -k night since 18 Rcl ? Bxc3 + 19 Rxc3 Na4 is a disaster and 1 8 0-0 loses ma terial to 18 . . . Bxc3 19 bxc3 N a4 . Best seems 18 Bf3 but after 18 . . . Rfd8 (or even 18 . . . Bxc3!?) I don't see tha t White is all t h a t well placed; certainly this w a s better t h a n w h a t hap· pens in the game. 1 6 Be6+ Kg7 17 Nd5! (Diagram 2 2)

A strong move. 1 7 0-0 Nexc4 was nothing like so clear. 1 7 ... Nxd5

17 ... Re8 i s met not by 18 b3? (when Black continues a s in the game but instead of . . . Nxc5 later has . . . b7-b6, as e 7 is defended) but by the

26

Main Line with 4

. . .

d xe 5

powerful 1 8 Qe2. Black c a n take the c· paw n but White castles a n d brings h i s rooks t o the cen tral fil es with great pressure. 18 cxd5 Qa5+ 1 9 b 4 Qa6 (Diagram 2 3)

Diagram 23

Diagram 24

Is a check on d3 a problem?

Can Black hold?

Perhaps Black was relying on this move, but Grischuk's calm reply makes it clear that the displacement of h is kin g i s not a s im portant as the consolidation of White's bind. 20 Q e 2 Nd3+ 21 Kfl Nxc5 22 bxc5 Qxe2+ Davies' 22

...

Qa5

might b e b e t t e r b u t Black stands poorly. White's

bishop on e6 is paralyzing both black rooks single-handedly, and I don't see how Black can rectify t his problem . Having said that, the ending arising after the text is surely lost.

23 Kxe 2 b 6

A deeply unpleasant move t o h ave to make, but otherwise the b·pawn can not be defended. 24 c6 B f4 25 K d 3 b 5 2 6 g3 B c 7 2 7 R h b l RibS 28 a4 b4 ( Diagram

24 )

Black's dark-squared blockade enables h i m to avoid losing on the queenside, but Grischuk o p e n s a second fro nt and the cra m ped black pieces can not defe n d both w i n gs. TIP: In the ending one weakness (here the queenslde) in the opposi ­ tion camp is seldom enough to win . You need to create a second weakness (here the kingside) and a lternate attacks unti l the defence cannot keep up.

29 Kc4 a5 30 Kb3 Bd6 3 1 R e l h 5 32 R a c l Ra7 33 Rc4 R c 7 34 R e e l Ra8 35 h 4 g5 36 g4 R h 8 3 7 R g l Kf8 38 gxh5 Rxh5 39 Rcg4 Kg7 4 0 f 4 Ra7 4 1 fxg5 Ra8 42 gxf6+ Kxf6 43 Rfl + Ke5 4 4 Rg5+ Rxg5 45 fxg5 1-0

Black's hel plessness towards the end was striking. A fine brisk win by

27

Starti n g O ut : A l e k h i n e ' s Defe n ce

Grischuk in h i s characteristic style, but I thi n k Black h a s more re­ sources in the opening.

Statistics White has scored 5 7% after 6 . . . Nb6.

Th eoretica l ? A bit, yes. 6 Qf3 is complicated rather tha n theoretical (and is a n i n ­ teresting field to a n alyse, i f you like t h a t k i n d of t hing) . One c a n play against 6 c 4 without k nowing a ny theory , but it's a dangerous ap­ proach, a nd worth a little study.

Theoret i ca l Status There h a ve n't been that many games , probably because this variation is sharp and White players suspect Black players will know more about it tha n they do. 9 Rc l has a good reputatio n , though no doubt there's much to be discovered. 6 Qf3 is supposed to be refuted, a l­ though I'm no t a t all su re.. . Black ca n also try to slide into the Ken gis by the move order 1 e4 N f6 2 e5 Nd5 3 d4 d6 4 Nf3 g6 5 Bc4 c6. I do not recommend thi s . White can reply 6 exd6 Qxd6 7 0-0 Bg 7 (7 ... Bg4 is not yet a good idea be­ cause of 8 h 3 Bxf3 9 Qxf3 Bg7 1 0 Nc3 !, exploitin g the weakness of f7) a n d now 8 h 3! leaves Black in breach of my three piece rule as he is u nable to trade a minor piece, which in turn mea ns he has to develop his c8-bishop to b7, for example 8 . . . 0 - 0 9 Re 1 Nd 7 10 Bb3 b5 1 1 Nbd 2 Bb7 12 Ne4 Q c7 13 Bg5 e6 was better for White i n Svidler-Akopian, Halkidiki 2 002 . It i s a lso possible that White can arrange to recaptu re with the d-pawn on e5, for example 6 h 3 !? dxe5 7 dxe5, the resulting set-up usually proving awkward for Black's kingside fia nchetto. Fur­ thermore, in comparison to lines wit h . . . e 7-e6 a nd . . . Be7, Black must either play ... e 7 -e6 a n d weake n f6, or spend the whole game worrying about e5-e6 fro m White . Game 4

0 A.David • C . Bauer Clichy 2004 1 e4 Nf6 2 e5 N d 5 3 d4 d 6 4 Nf3 dxe5 5 Nxe5 c6 ( D i agram 2 5)

Based o n the remarks i n the introduction, you wouldn't expect 6 Bc4 to be critical, and it isn't. A good e x a m p l e is va n der Weide-Miles, Saint Vincent 2 000, which con ti nued 6 . .. Nd 7 7 Nf3 (7 Nxd 7 Bxd 7 8 00 Bf5 is complet ely equal, for example 9 Bg5 Qd 7 1 0 Bb3 e6 1 1 c4 Be 7 !) 7 ... N 7b6 8 Bb3 Bg4 9 Nbd 2 e 6 10 0-0 Be 7 1 1 h 3 Bh5 (this is the difference from the Kengis: Black keeps the bishop , which is consid­ erably more effective tha n White's, which is stuck on b3) 12 c3 0-0 13 Ne4 Bg6 14 Qe2 Nd7 15 Re 1 a5 16 a4 Re8 (preparing . . . e6-e5) a n d Black was very comfortable.

28

Main Line with 4

.

.. d x e 5

Dia g r a m 2 5

Diagram 26

T h e flexible 5 . . c6

Black's trademar k challenge

.

6 Nd2 seems wrong to me. White tries to maintain a k n ight on e5, but exchanging a pair of minor pieces must surely help Black. Black has usua l ly played 6 . . . Nd7 7 N df3 (7 Nd3 is illogical, p erh a ps Black's sim­ plest being 7 ... e6 and ... c6-c5) 7 ... Nxe5 8 Nxe5 g6 9 Q f3 ( this strong placement of the queen is one of White's compensations for having al­ lowed the undesirable e xch a nge of knights in Nd2 systems) 9 . . . Be6 1 0 c 3 Bg7 1 1 Be2 a5, leading to positions very similar to the Kengis sys­ tem. Black will probably find himself forced to move his bishop from e6 and exchange it for the knight on d3, when White can always claim a tiny edge with his two bishops (the unopposed light-squ ared bishop can be a particular demon). I ns te ad I ra ther like Black's play in Zapata­ Grunberg, Isle of Man 2002, where Black played 8 . . . Bf5 instead of 8 . . . g6. After 9 Nx£7 Kxf7 10 Qf3 Black can draw at once with 10 ... Kg6 1 1 Qg3+ Kf7 12 Qf3 or play to hang on to the piece with 10 . . . Ke6 1 1 Bc4. H e should avoid 1 0 . . .e 6 1 1 g 4 Nb4, which a t first sight looks quite strong, but actually after 1 2 gxf5 Nxc2+ 1 3 Kd 1 Nxa 1? just loses to 14 fxe6+, so Black has to continue 13 . . . Nxd4 14 fxe6+ Kxe6 15 Qg4+ N f5 +, when Black's exposed king is the more serious. White, too, has to be a little carefu l after 6 Nd2 N d7 7 Ndf3 Nxe5 8 Nxe5 Bf5 , since the routine 9 Be2 e6 1 0 0 -0 Nb4 is decidedly embar­ rassing, while 9 Bd3 Bxd3 10 Qxd3 concedes equality at once. Zapata­ Grunberg itself quickly lurched out of control with 9 Qf3 e6 10 Bd2 (and not 10 Bd3? Bb4+ 1 1 Kd 1 when White's inadvertently wandering king proved too big a handicap in Arakhamia-Baburin, 4 N CL 2004) 10 . . . Qb6 1 1 0-0-0 Qxd4! 1 2 Bd3 Ba3!? (did White really have enough for his pawn after 12 . Qxe5 1 3 Rfe l Qf6 and so on?) 1 3 Nc4 Bxd3 (White was offering the exchange after 13 . .. Bg4 1 4 Qg3 Bxd l , when it's not clear to me tha t either 15 Rxd l Be7 or 1 5 Nxa3 Bh 5 offer enough com­ pensation) 14 Nxa3 Bf5 15 c4 Bg4 16 Qg3 Bxdl 1 7 Rxd l . This time 17 ... Nf6 18 Bb 4 is too dangerous, and both players chose to bale out with 1 7 ... 0-0-0 1 8 Bg5 Qc5 19 Bxd8 Rxd8 20 Qxg7 Nb4 with equality. . .

29

S t a rt i n g O u t : A l e k h i n e ' s D e fe nce

6 g 3 N d 7 7 Nf3 transposes t o Game 9 . 6 Bd 3 i s the next game . 6 Be2 Bf5

This move is a recent invention, although very logical. Black wants to get in . . . Nd7 without the inconvenience of having to move the knight again to develop his bishop if White replies Nf3. The older - and still very valid - method was 6 ... Nd7 7 Nf3 g6 8 c4 Nc7 9 Nc3 Bg7 10 0·0 0 0 , which Tony Miles played a number of times. Then 1 1 Re 1 c 5 1 2 d 5 b 5 (de Firmian-Miles, Chicago 1994), 1 1 Bf4 c 5 12 d 5 (Degraeve-Miles, Mond ariz 2000) or 12 dxc5 (Watson-Baburin Kilkenny 1994) an d 1 1 Be3 a 6 1 2 a 4 a 5 (Apicell a-Miles Linares 1 995) all give Black a decent position. White might do better to delay 8 c4 and instead play 8 0-0 Bg7 9 Re 1 0 · 0 10 Bf 1 with the idea of meeting 10 . . . Bg4 with 1 1 N bd2 and h2-h3. This position can also arise in the Kengis variation, and Black should continue as there with ... Bxf3 and ... e7 -e6 and prepare . . . c6-c5. 7 0-0

The main alternative to this is Kasparov's brash 7 g4, which has also been the choice of Almasi and Judit Polgar Black re pli e s 7 . . . Be6, and now not Kasparov's 8 c4 Nb6 9 b3 f6 w h en Black wins a pawn, but 8 f4 f6 9 Nd3 (or 9 f5 Bc8 a n d . . . e 7 -e6) 9 . .. Bf7 10 0-0 Na6, when Black seems fine a fter 11 Nc3 Nxc3 1 2 bxc3 g6 (Polgar-S hort Bled 2002) or 1 1 Nf2 Nac 7 , although both positions are messy and unclear. .

7 ... Nd7 ( D iagram 26) 8 B g4

Rather feeble. The natural way to play is 8 Nf3, when Loeffler's clever 8 ... h6 is probably a good choice. The idea of this is to make a useful move while preventi n g 9 c 4 , when 9 ... Nb 4 10 a 3 Nc2 1 1 Ra2 Nxd4 now wins a pawn (had Black pl ayed 8 . . . e6, then 12 Qxd4 Bxb1 13 Bg5 would be strong)_ A continuation like 9 a 3 e6 1 0 c 4 N 5 f6 1 1 Nc3 Bd6 12 b4 0 - 0 1 3 Bb2 a6 14 Re 1 b5 was fine for Black in Luther- Loeffler, Austria 2002. Notice the active dep l oy m ent of the bishop on d6 ra ther than e7, and the . . . a 7-a6 a n d . . . b7 - b 5 m anoeuvre - both are typical of Black's play in this l ine. 9 Bd3 Bxd3 10 Qxd3 e6 11 c4 N 7f6 1 2 Bf4 Be7 1 3 Rad1 Qb6 is a lso equal (Ramesh-Cox, Isle of Man 2004). 8 ... Bxg4 9 Qxg4 e 6

White was hoping his queen would exert some a nnoying pressure on g7 a fter 9 ... Nxe5 1 0 dxe5 . Th at w asn't so bad fo r Black, but the text is easier. 10 c4 Nb4 1 1 Nxd 7 Q x d 7 1 2 Qe4 Rd8 13 a3 Na6 (Di agram 2 7 )

White manages t o drive Black back, b u t h e h a s no advantage . 1 4 B e 3 Be7 1 5 N d 2 Perhaps 1 5 Nc3 w a s more natural , but b y threatening t o bring the knight to e5 White manages to provoke a slight weaknes s .

15 . 0-0 16 Nf3 f6 17 Rfe 1 Rfe8 1 8 B d 2 Nc7 1 9 Ba5 b6 ..

White would rather like to exc h ange his bishop for the knight o n c7, since t h e latter both defen ds e6 a n d prevents all hope o f a d 4 - d 5 break , while supporting Black's o w n potential pus h o f the b -pawn.

30

Main Line with 4

...

dxe5

Naturally B lack prevents this.

Diagram 27

Diagram 28

Black i s fine

Black u ndertakes positive action

20 Bc3 Bf8 21 Rac l R e S 2 2 h4 h5 2 3 R e 2 Bd6 24 Rcc2 Bf8 2 5 R e d 2 b 5 (Diagram 2 8)

Neither side has m uch to do, but after due consideration Ba uer de­ cides he is not satisfied with a draw by shuffling. Ti P : In this pawn structure, Black's pri mary breaks are .. .c6-c5 and ...e6-e5 , b ut pushing to bS to re move the c-pawn and estab lish an uncha l lengeable out post on dS is a l ways another option.

26 Ba5 Na6

Of course not 2 6 ... bxc4 2 7 Bxc7. 27 Re2

A curious m ove which ra ther i n v i t e s Black's next. Something like 27 Qe2 was sensible. 2 7 ... c5

With the k n ight o u t of range of d5 27 . . .bxc4 is no good. White might h a v e been advised to try 28 Red2 now, when 28 . . . cxd4 29 R x d 4 Qc6 30 Qxc6 R xc6 3 1 b4 e5 32 Rd 7 Re7 33 Rxe7 B x e 7 34 c 5 is a sam ple con­ tin uation to show B l a ck's ideas - if he ca n take the knight to c6 and king to e 6 he will be better. After 28 cxb 5 Qxb5 29 Bc3 c4 Black ca n use d5 for his knight after all. 28 dxc5 Nxc5 2 9 Qg6 ( Diagram 29)

I t looks as though time trouble was affecting the play. The text is not bad, but it needs following up correctly. 29 ... bxc4 30 Qxh5

3 0 Rxc4 was necessa ry. The n neither 30 ... Qb5 3 1 Rec2 Qxa5 32 b4 Qxa3 33 b x c 5 , nor 30 ... Q d 3 31 Qxd3 Nxd3 32 Rxe6, is qu ite as good as i t looks for Black, so the position is still about equal.

31

S t a rt i n g O u t : A l e k h i n e' s D e fe nce

30 . . .N d 3 (Diagram 30)

Now , however, Black is doing well with his strong knight, a n d to m a k e matters worse White notices only one of the threats.

Diagram

D i ag r a m 29 The temperature r ises

30

Spot Black's threa t (s)

3 1 Re4?

To prevent 3 1 . . . Nf4 , but Black h a s another idea. 3 1 Qg4 i s be tter. 3 1 . .. Rc5 32 Qg6 Rxa5 33 Rexc4 Ne5 3 4 Nxe5 Rxe5 35 Rc7

White m a y have experienced a glim mer of hope at this point - espe­ cially if he saw 35 ... Re 1 + 36 Kh2 Bd6+?? ( 3 6 ... Qd6+ is still good enough) 37 g3 Bxc7 38 Rxc7 Qxc7 39 Qxe8+ with perpetual - but Bauer soon extinguishes it. 35 . .. Bd6 36 g3 0-1

36 Rxd7 Re l mate, of course.

Statistics A t the moment Black h a s scored about 60% after 6 . . . Bf5.

Th eo retica l ? No.

Theo retical Status Very good. White doesn't y e t have t h e answer. Game S

0 E.Su tovsky • M.Carlsen Rethymnon 2003

1 e4 Nf6 2 e5 Nd5 3 d4 d 6 4 Nf3 dxe5 5 Nxe5 c6 6 B d 3

A somewhat awkward looking move, b u t White wa nts t o prevent the

32

M a i n Li n e w ith 4 ... dx e5

development of the queen's bishop. 6 . . . Nd7 7 0-0

Mter 7 Nf3 I thi n k Stuart Conquest showed the w a y i n d e l R i o An· gelis- Conquest , Ista nbul 2 003, with 7 . . . N7f6 8 h3 (otherwise 8 . . . Bg4) 8 . . . Nb4 9 Bc4 Bf5 , when 10 Ne5 e6 1 1 Na3 Bd6 12 0·0 h6 1 3 c3 N bd5 14 Nc2 Qc7 w a s about e qual. 7 Nxd7 Bxd7 (recently given the Mickey Adams' seal of a pp rova l) makes sense inasmuch as it prevents this manoeuvre, and after 8 0-0 g6 9 Re 1 Bg7 1 0 c3 0-0 1 1 Bg5 Re8 12 Nd2 Qc8 somethin g akin to a quiet line of the Kengis system arises. Ste­ fansson-Gausel, Aarhus 2003 continued thematically with 13 Nc4 (better tha n the previously played 13 Ne4) 13 . . . Bf5 14 Be2 Qc7 1 5 Bf3 Rad8 1 6 Qb3 when White has a bind a nd is pla n n i n g to i ncrease the pressure with Rad 1 and Bh4· g3. Black played the provocative 16 . . . Be6 and White was i ndeed provoked: 1 7 Rxe6 fxe6 1 8 Re1 (Dia­ g ram 3 1)

Diagram 31

Diagram 32

A thematic exchange sacrifice

An early queen trade

TI P: In the Kengis pawn structure Rx(B)e6 is always worth consider­ ing.

It is worth tryin g a few ways to free Black's position. The sacrifice is n 't i nte nded to quickly deliver mate or a nything, rather to obtain a positional bind. I t is surprisingly difficult to obtain li nes for the rooks and/or d e fend the pawns on e6 a n d b 7 , while the pawn on e7 cuts Black's position in half and m a kes it very difficult to defend the king. I thi n k White stands c on side ra b ly better here, as the rest of the game demonstrates. Notice how little Black fi nds to do beyo nd creating more weaknes ses as White regroups his pieces: 18 ... b5 19 Ne5 Rf8 20 B g4 Rd6 2 1 N d 3 h 6 22 Bh4 Kh7 2 3 Bg3 e 5 2 4 dx e 5 Rdd8 2 5 Nc5 Q b6 26 Ne6 h 5 27 Bx h 5 gxh5 2 8 Qc2+ Kh6 2 9 Bh4 Bf6 30 exf6 e x f6 31 Qf5 Rg8 32 Re5 1 -0 . Re tu rn in g to t h e op e n i n g , i t seem s to me t h a t 1 4 . . . b 5 w a s the move.

33

S t arti n g O ut : A le k h i ne ' s D e fe n ce

1 5 Ne5? f6 wins a piece, while otherwise t h e t hematic . . . e 7 - e 5 equal­ easily, for example 15 Ne3 Nxe3 16 Bxe3 e5.

ises

7 . Nxe5 .

.

This was not essential: 7 ... g6 is a reasonable way to play, whe n 8 Nf3 is an inferior version of the note to Black's sixth move in Game 4 and 8 Nxd7 is Stefansson-Gausel . 8 d x e 5 Nb4 9 Be4 Qx d 1 10 Rxd 1 (Diag ram 3 2) It's a l ittle u n usual for White to allow a n endgame like this i n the Alekhine - or at least to get anywhere when he does - a n d the idea that White could be seriously better will raise the hackles of a n y Alekh ine fan . I su spect Sutovsky's choice of opening h a d less to do with objectivity and more to do with the fact that Carlsen is a 1 3 year-old tactical genius, a nd against that kind o f opponent grandmas­ ters t end to reckon that the sooner you ca n get the quee n s off and ap­ ply a bit of positional n ous a n d experience the better. 1 0 £5 ...

A good move, shifting the bishop from its ideal position and gai ning both space a n d time (the bishop will n eed to move yet again to allow f2 - f4) . 1 1 a 3 Na6

1 l . . .fxe4 1 2 axb4 was certainly conceivable and perhaps best. After the line M a rtin gives on chesspublishi n g.com, 12 g6 1 3 Be3 Bg7 1 4 R a 5 a 6 1 5 N c 3 Bf5 1 6 h3 h 5 , Black h a s h i s own play. I n deed i t is n ot so simple, e.g. 1 7 Rd4 (to collect the e-pawn) 1 7 . b5 1 8 Bf4 e3, while Black has the idea of Kf7 -e6. . . .

. .

. . .

12 Bf3 g6 13 Nd2

At first sight it would have been better to play 13 Be2 immediately. If White is going to play Bxa6 bxa6 at any poi nt, then he wa nts to follow up with Nc3 since in that way he ca n both keep the black king away from the queenside and s e n d the knight to c 5 without havi ng to go via b3 (where the bishop may capture it). 13 Be2 Nc7 14 Nd2 Be6 is the game, but it avoids the opportunity I think Black had at move 1 4 , while 13 Be2 Be6 1 4 Bxa6 bxa6 1 5 Nc 3 (-a 4-c5) would worry m e i f l w a s Black. But this is a very di fficult position and I am sure Sutovsky had a good reason for his move order. As it happe n s I was in Rethym­ non watch ing this game a nd Sutovsky thought for a very long time over t h is a n d subsequent moves up to move 2 0 or so; thereafter h e rattled his moves out v e ry confidently. 1 3 ... Be6 14 Be2 ( D iag ram 3 3) TIP: In this kind of ending Black a l most always castles queenside,

where h is k ing can defend the pawns and his rooks c an g et to the d­ fi le and/or support a break on the kingside with

...

g6-g5.

Call me a result merchant, but in the e nd Black loses this game be­ cause his king gets stuck on the kingside while White queens his a ­ paw n , a n d i f i t wasn't for the fact that I a m a humble hack while Mas-

34

Main Line with 4

•..

d xe 5

ter Carlsen m a y b e t h e next world champion b u t two I would say 14 0-0-0 w a s downright m andatory here. As i t is I merely put i t for­ ward as a suggestion. After that I don't think Black stands so b a dly. If W h i te goes o n with 15 Nf3 as he does in the game then 15 . . . Rx d1+ 1 6 B x d 1 Bd5 1 7 N d 4 B g 7 1 8 f4 Rd8 1 9 Be3 N c 7 is far better than what h a p pens. I can't think why Carlsen rej ected this. It may have been 1 5 B x a 6 b xa6, for example 1 6 f4 Bh6 1 7 Rfl Rd7, but it seems to m e that this m u st be okay for Black tha nks to his strong li ght squared b ishop. ...

­

Dia g ra m 3 3

D i a g ra m 34

How would y o u play?

The bishop

is forced

back

14 . . . Nc7?! 15 Nf3 Bd5 1 6 Nd4 Bg7 1 7 c4! (Diagram 34) This is th e difference: White has th e tempo to do this before he needs to defend h is e-pawn, so tha t the bishop cannot come to e 4 i n view of f2-f3.

1 7 . . . Bf7 1 8 f4

Now White h a s his ideal position a n d Black is really in trouble. 1 8. . . N e 6

I would still have preferred 1 8 . 0 - 0 - 0 myself, but by n o w I m a y be re­ sulting it a little; no doubt Su tovsky would still have managed to cre­ ate real problems for Black. .

.

19 Nf3 Rd8 2 0 Be3 0-0 2 1 b4 a6 22 g3 Rxd1 23 Rxd1 Rd8 (Dia­ gram 3 5 )

Before B b 6 wins the queen's fil e, b u t the price Black ha s pa id to ex­ change the rooks - his king and both bishops being out of p l ay is too much. -

24 Rxd8+ N xd8 25 Nd4

So simple . Wh ite w a n ts to win the b-pawn with Nb3, N a 5 and Bb6. Black is a l re a dy gone. 25 ... Be8 26 Nb3 B d 7 27 Nc5 Bc8 2 8 b 5 ! (Diagram 3 6) 2 8 . . . c x b 5 29

35

Start i n g O ut : A l e k h i n e ' s D e fe n c e

c x b 5 axb5 30 B x b 5 Bf8 3 1 Kf2 Kg7

Diagram 36

Diagram 3 5 Exchanges fail

to h e l p Black

The decisive break

Since 3 1 . . .e6 loses to 32 N a 4 Nc6 33 Nb6 N a 7 34 Be8 followed by Nxc8 and B d 7 , Black cannot get his king's bishop and king into the game and loses on the queenside .

3 2 N d 7 � 3 3 a4 Bxd7 3 4 Bxd7 e6 3 5 B b 6 Be7 3 6 a 5 g5 3 7 Ke2 h5 38 Bc8 h 4 39 B x b 7 1-0

The a -pawn queens. An instructive ga m e a n d possibly an im portant variation.

Statistics White has scored 3 1� /4 with this particular idea (7 0-0) thus far.

Theo retical? No.

Theoretica l Status Sutovsky's idea i n t h i s game is the only problem I s e e at t h e moment, but really there's too little theory yet to talk about theoretical status . Game 6 0 Adams • Burkart

European Club Cup 1 996

1 e 4 Nf6 2 e 5 Nd5 3 d4 d 6 4 Nf3 dxe5 5 Nxe5 c6 6 c4 Nb4!? ( D i a­ gram 3 7)

A rem arkable move, also m ade possible by 5 c6, which prevents Qa4+ fro m coll ecting a piece. There hasn't been a tremendous a m ount of experience with it yet, but such l uminaries a s Tim m an, A da ms a n d Baue r have failed t o show anything with White, so i t deserves a look. . . .

36

M a i n L i n e w i t h 4 . . . d x e5

There's nothing a t all wrong with 6 ... Nc7, which will probably lead to lines similar to those considered in the note to Black's sixth in game 4.

Diagram 38

Diagram 37 Black invites 7

a3

White forces t h e issue

7 Be3

7 a3? loses to 7 . . . Qxd4 8 Nxf7 Q e4 +, a n d 7 N d 3 e5! is also very com· fortable for Black (8 Nxe5 Qxd4) , e.g. 8 dxe5 B f5 9 N xb4 B xb4+ 1 0 B d 2 Q a 5 , when 1 1 f4 ? gives B l a c k m uch too m uch play after some­ thing like 1 l . . . N a 6 12 Nc3 0-0-0, while 1 1 Bxb4 Qxb4+ 1 2 Qd2 Na6 1 3 Be2 ( 1 3 a 3 Qe7, a gain followed b y . . . 0-0-0 a n d . . .Nc5) 1 3 . . .Rd8 1 4 Qc3 0 - 0 1 5 0-0 Rfe8 1 6 a3 Qc5 m a d e no attempt to defend the e-pa w n a n d acknowledged tha t Black w a s a l i ttle better (if anything) i n Brodsky­ Loeffler, Hamburg 200 1 . 7 . . . B f5 8 Na3

Thus far the opening h a s b e e n a considerable success for Black. The a 3-square is not where the kn ight wanted to be stabled, a n d Black's knight on b 4 now fee ls m uch more secure. There is no real alterna­ tive, though, since 8 Nd 3 e5 9 N xb4 + B xb4 1 0 Nc3 0-0 achieves zilch a fte r 1 1 dxe5 Nd7 (Timman-Luther, Sweden 2002) or 1 1 Be 2 cxd4 1 2 Qxd4 N a 6 (Bauer-Loeffler Lausanne 200 1 ) . 8 . . . N d 7 9 Nxd 7 I imagine you know the answer t o 9 N d3 ( o r 9 Nf3) b y now, nam ely

9 . . . e5 1 0 dxe5 Qa5 which is unpleasant for White, but the text gets nowhere either.

9 Qxd 7 1 0 Be2 g6 .••

As Davies points out 1 0 . . . e6 a lso looks j olly reasonabl e. 1 1 d5 Bg7 1 2 Bd4 (Diagram 38)

Th i s is really White's only try for a n advantage. B lack has a sim plify. ing reply but it does at least allow White to get rid of his spavined n a g

37

S t a rt i n g O u t : A l e k h i n e ' s D e fe n c e

on a 3 . 1 2 . . .Nc2+ 1 3 Nxc2 Bxc2 1 4 Q d 2 Bxd4 1 5 Qxd4 0-0 1 6 0 - 0 cxd5 1 7 cxd5 Rfd 8

F o r someo ne with t h e Black pieces playing a n opponent rated 300 points above h i m , Burkart isn't doing a t all b a d ly, but the text is a slight and i nstructive inaccuracy . Davies' 1 7 Qd6 was better since it wou l d have enabled B lack to meet 18 Rac l with 1 8 . . e5! , after which the passed d-pawn is as much of a weakness as a strength, and Black avoids the pressure on the e-file that occurs in the game. Black is ab­ solutely equal then, whereas in th e game Ada m s maintains a l ittle edge. . . .

.

1 8 R a c 1 Rac8 1 9 Bf3 b 6

Black needs t o attend t o the a-pawn now, which gives White time to c l a m p down on the e -pawn. 20 Rfe 1 (Diagram 39)

Diagram 3 9

Diagram 40

White exerts pressure on the e-fil e

The tension m o unts . . .

20. . .Q d 6 2 1 h 4 B f5 2 2 Rcd 1

And not 22 h5?, when B lack has 2 2 . Rxc 1 2 3 Rxcl e5 again . . .

22 . . . h5 23 Re3 R d 7 2 4 R d e 1 Rdc7 25 B d 1

White im proves h i s worst piece a n d i n passing prevents Rc2, but he is plunging into co m plications which I think fa vour Black objectively. 25 g3 was a sensible a lternative. ...

25 ... Rc4 26 Qd2 e 6

Not 2 6 Rxh4 2 7 g 3 a n d Rxe7, but now the position b ecomes sharp. . . .

27 Re5 Rxh4

A bold move, but a good o ne. 2 7 . . . Rc l was certainly possible and ties Wh ite's forces down to a large extent since 28 Bb3? is met b y 2 8 .. Qxe5. I don't know quite how White would have proceeded then, but 27 . . .Rxh4 i s better. .

38

M a i n L i n e w i t h 4 ... d xe 5

2 8 Bb3 Q b 4

T h i s is fine - i n fact i t forces a draw - but to be honest I ca n't se e wh at Mickey A d a m s was going to p l a y after 28 . . . f6. Accord i n g t o one of those horrible m achines, 29 Rxe6 simply loses to 29 . . . Qh2+ 30 Kf1 B g4 when defending is hopeless, for exa m ple 31 f3 Bxf3 32 Qf2 Bxg2+ 33 Ke2 B f3+ 34 Ke3 Qf4+ and so forth, a n d there is no perpetual after 31 Re8+ Rxe8 3 2 Rxe8+ Kf7 33 d6+ Kxe8 since 34 Qe3+ is met by 34 . . . Qe5. The heartless beast reckons that 29 Rxf5 Qh2+ 3 0 Kf1 Qh 1 + 3 1 K e 2 Re4+ 3 2 Kf3 Qxe1 wins too since 33 Qh6 a llows m a te with 33 . . . Qe2+, . . . Qg4+, . . . Qh4+ and . . . Re 1. 29 R5e3 Qh2+ 30 Kf1 Qh 1+ 31 Ke2 Qxg2 is also grim, for example 32 dxe6 Bg4+ 33 Kd3 Rd8+ 3 4 Kc3 Q c 6+ 35 Bc4 Qxc4+ 36 Kxc4 Rxd2 37 e7 Bd7+ 3 8 Kc3 Rx£'2. This only seem s to leave 29 g3 fxe5 30 gxh4 Kg7, but I c a n 't see that W h ite has a n a dequate contin uation here. 29 Qh6 Be4

The only move; no doubt Bl ack had been relying on it. 30 Re2! (Diagram 40)

3 0 R d 1 i s weaker: 30 ... exd5 31 Bxd5 Bxd5 32 Rex d 5 Re8 leaves Black a pawn up as after 33 Re5 the cool 33 . . . Rh 1+ 34 Kxh 1 Qh4+ 35 Kg l Rxe5 defends d8. After the tex t 30 ... exd 5 3 1 Bxd 5 Bxd5 ?? al lows m a te in three with 32 Re8+, although even here instead of 3 l . . . Bx d5 t h e continuation 3 1 . . . Qd6! 32 Bxf 7+ Kxf7 33 R2xe4 R x e 4 34 R x e 4 Kg8 s eems to hang o n . B u t probably the most expedient course w a s t o force a draw with 30 . . . Rh 1+ 3 1 Kxh 1 Bxg2+ 32 Kxg2 (not 32 Kg1 ? Qg4 winning, for ex­ a m ple 3 3 R/Qg5 Rc 1 +, or 3 3 f3 Bxf3+ 34 Kf1 Qh3+ 3 5 Ke 1 Qh 1 + 36 Kf2 Qg2+ 37 Ke 1 Qg l+ 38 Kd2 Qc l+ winning the queen) 32 ... Qg4+ 3 3 Kfl Qh3+, a n d White must a llow perpetual since 3 4 Ke 1 Q h l + 3 5 Kd2 Qc l+ 3 6 K d 3 Qxh6 3 7 d xe6 Rd8+ 3 8 Kc2 fxe 6 3 9 Rxe6 Kh8 loses the queen for i n a deq u a t e compensation . I n s t e ad Burkart falters after his hitherto excellent play, either in time trouble or overlookin g some­ thing very sim ple. 30

..•

Rg4?? 3 1 dxe6 Rxg2+ 3 2 Kf1 Bd5 33 exf7+ 1-0

An interesting exa mple of how a n unexpectedly strong quiet move like 30 Re2, even if it isn't objectively so devastating, can cause a rea ­

sonable player to co llapse.

Statistics Black has scored 5 5 % after 6 . . . N b 4 .

Theoretic a l ? Not really. J u s t rem e mber to meet N d 3 with . . . e7-e5 and you'll b e fi n e.

39

Start i n g O u t : A l ek h i n e ' s D e fe n ce

Theoretica l Stat u s Exce ll ent. The ball is firmly i n White's court a n d since t h e early play is quite forcing it won't be easy to find anything. I won't be surprised if 6 c 4 comes to be seen as a mistake.

That ends our coverage of the Miles variation which, at the ·moment, is one of the very best ways for Black to play against the Main Line. White has to t hink for himself and as yet those thoughts haven't done terribly well against any of Black's options presented here. It is also possible to introduce t he system by 4 ... c6, planning 5 Be2 dxe5. This cuts down White's options a bit since it reserves the possibility that Black's next move will be 5 . . . Bg4 , tran sposing to the system with 4 . . . Bg4 and 5 . . . c6 (see page 5 1 of Chapter 2). For example, if the Bd3 idea of Game 5 bothers Black, then this is an excellent device, since 5 Bd3 Bg4 6 h3 Bxf3 (or 6 . . . Bh5, indeed) 7 Qxf3 dxe5 8 dxe5 e6 is not that good for White; the same is true of 5 Bc4 (or 5 g3) . There are various drawbacks, though. First of all the handy 6 . . . Nb4 of Game 6 is not available, so after 5 c4 Nc 7 6 Be2 Black would be committed to the older lines mentioned in the notes to Ga m e 4 (or to 6 . . . Bg4 with a direct transposition to Chapter 2 ) . Secondly, White retains the option of 5 Be2 dxe5 6 dxe5 when Black should avoid 6 . . . Bf5 7 Nd4! (7 . . . Bg6 8 e6, or 7 . . . e6 8 Nxf5) in favour of 6 . . . Bg4 . 7 c4 Nb4 8 Qxd8+ is com­ fortable for Black, and so is 8 Qb3 Bxf3 9 gxf3 Qb6. However, I'm not too s u re what Black's next trick is after 7 0 - 0 . Of course he c a n trans­ pose into the 4 . . . Bg4/5 . . . c6 system with 7 . . . Bxf3. But he should beware of 7 . . . e6 8 Ng5 which is a n old variation of the 4 . . . Bg4/5 . . . c6 syste m co n sid ere d unpleasant for Black (the queen is perfect on e2, the knight on g5 can come dangerously to e4 a nd d6 and White gains by delaying c2-c4 since the other knight can sometimes come to that square) . Black would like to preserve the bishop with 8 . . . Bf5 but 9 c4 is very strong (9 . . . Nb4 1 0 Qxd8+ Kxd8 1 1 Nxf7+ Ke8 12 Nxh8 Nc2 1 3 g 4 etc just lo s e s , s o the grim 9 . . . Be7 1 0 cxd5 Bxg5 1 1 f4 B h4 1 2 d 6 is forced). Thirdly, White has the possibility of moving into a type of Ex­ change variation, as seen in the next game. Game 7 0 Gal lagher • Bab urin Isle of Man 2001

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 c6 (Diagram 4 1 ) 5 c4 Nc7 6 exd6 exd6 7 Nc3 Bg4 8 h3 B h 5 9 Be3 Be7 1 0 g4 An aggressive response, although , of course, White could play more

quietly with play similar to Games 26 a n d 2 7 . 1 0 . . . Bg6 1 1 Q b 3 Qc8

1 l . . .b6 12 Bg2 0 - 0 13 h4 h 5 14 gxh5 Nxh5 15 Ng5 gave White a n attack in McShane-Baburin, Kilkenny 1 998, but since then Baburin has de­ vised an entirely new plan, of which this move is on ly the b e ginn in g .

40

Main Line with 4

dxe5

Diagram 42

Diag ram 41 A diffe re nt

...

move o r d e r

A

useful waiting move

12 Be2

With c6 not weakened 1 2 .. .a 6

1 2 Bg2

has less point.

(Diagram 42)

Black delays castling until he sees where Wh ite ca stles, and also pre­ pares . . . b7-b 5 . 13 Nd2

1 3 h 4 has no point u n t i l Black the push of his f-pawn.

castles

on t h a t side, so Wh ite prepares

13 ... Nd7 1 4 0-0-0

This doesn't work out at a l l well, though it's n o t clear it was bad yet. Flear's 14 f4 B h4 + 15 Bf2 Bxf2 16 K x f2 was perhaps preferable, but I don't think Black has m uch to worry about after 1 6. . .f5 (much better than 16 .. .f6 as Black m ust keep control of the e4-square; otherwise a knight appears there with gre a t force a fter 16 .f6 17 f5 Bf7 1 8 Nde4). The key to the opening was to avoid premature ca stling and weaken­ ing the light squares with ... b 7 -b6. . .

14 ... d5 15 f4 f6

Since Black has h a d time to play . . . d6-d 5 , controlling e4 that way, h e no longe r needs t o p l a y t h e w eak en in g . . . f7 -f5. 16 f5 Bf7 White's driving the bishop into the attack without actu ally achieving much himself. I'm sure Joe Gallagher a l rea dy had a bad feeling in his waters, b u t it's difficult to say what he should h a v e done.

1 7 K b 1 b5 1 8 cxb5

1 8 c 5 N xc5 18

.

19

dxc5 d4 was worse.

. a x b 5 1 9 Bf4 N c 5 ! ? (Diagram 4 3 ) .

41

S t a rt i n g O u t : A l e k h i n e ' s D e f e n c e

Diagram 4 3

Diagram 44

A n i c e move

Good bishop hunting

I doubt if this came as a shock to Gallagher, although it does to my com pu ter. To be fai r, the thing does try to fight back with 20 dxc5 d4 21 Nxb5 , when 2 l . .. Bxb3 22 Nxc7+ Kf7 23 N xb3 is dangerous for Bl ack, and neither 2 1 . . . Nxb5 22 Bc4 nor 2 l . . . cxb5 22 Bxb5+ Nxb5 23 Qxb5+ Kf8 are a t all clear. Obviously White has a n arduous defence ahead of h im , but especia lly if conducted by a com puter there isn't n ecessarily any reason to despair. I would have thought this was pre fe r abl e to the game, though , i n which White is lost within a fe w moves, although I am sure Gallagher a p p re ci ated that and his deci­ sion to steer clear of th ese com plications must be respected . 20 Qc2 b4 2 1 Nb5!?

2 1 dxc5 bxc3 2 2 Qxc3 Rxa2 i s scary (23 Kxa2 d4+), a l though 23 Bd6 keeps White fighting. 2 1 ...Nxb5 22 Bxb5 cxb5 23 dxc5 0-0

Gallagher's heroic m easures have at least kept most of the lines closed, although the trouble is th a t White has n o m aterial solace for his difficulties. 24 Rhel Bxc5 25 Nb3 d 4 ! 26 Qxc5

26 N xc 5 Bxa2+ 2 7 Kc l b 3 w ins. 26 ... Bxb3 2 7 QxcS Bxa2+ 2 S Kcl RfxcS+ 2 9 Kd2 Bd5 3 0 Kd3 RdS 3 1 Ral R x a l 3 2 Rxal Bg2 ( Diagram 4 4 )

The s m o k e h a s cl eared a nd Black h a s a decisive advantage, although - as ever - Gallagher fi ghts h a rd. 33 h4 Bh3 34 Rgl h5 35 gxh5+ Bxf5 36 Kd2 Kf7 3 7 Rcl Ke6 3 S R c 5 R h S 3 9 h 6 g x h 6 4 0 R x b 5 h 5 4 1 Rb6+ K f7 4 2 Rxb4 R d S 4 3 Rb7+ Kg6 4 4 R b 6 R e S 4 5 R d 6 d 3 4 6 B g 5 RfS 4 7 Be3 R e S 4 S Bg5 Rc2+ 49 Ke3 R e 2+ 5 0 Kf3 Be4+ 51 Kg3 Kf5 52 Rxf6+ Ke5 53 Rf2

This clearly wasn't i n the programme and somewhere Baburin m ust

42

M a i n l i ne w i t h 4 . d x e 5 ..

h a v e b e e n a b l e t o p l a y more accurately. White h a s genuine drawing chances now, although I dare say both sides were short of time by this point. 53 Re i 54 Bf6+ Kd5 55 K f4 Bg6 56 Bg5 Re4+ 57 Kg3 Rg4+ 58 Kh3 •..

I don't know w h y Gallagher rejected t h e more natural 5 8 Kf3. Threat­ ening mate by 58 . . . Kd4 can then be dealt with by 59 Bf6+ and as far as I can see White has very good chances to hold. I t might have had something to do with having seconds left to finish the game, of course - maybe h3 was j ust nearer the clock than f3. 58

...

Ke4 5 9 Bf6 K e 3 6 0 Rfl Ke2 6I Rh I Be4 0-I

62 Rh2+ Bg2+! wins a t once.

Statistics Hard to s a y : the transpositions blur t h e numbers.

Theoretica l ? Not really, but the ideas i n this game are i m portant.

Theoretica I Status I don't think t h i s Exchange-type syste m is too terrifying. I do t h i n k 5 Be2 dxe5 6 dxe5 is dangerous, unless of course Black is happy to transpose i nto the 4 . . . Bg4/5 . . . c6 syste m . Game S 0 Kal lai • C.Horvath Budapest 1 990 I e4 Nf6 2 e 5 Nd5 3 d4 d6 4 Nf3 dxe5 5 Nxe5 Nd7 (Diagram 45)

D iag r a m 4 5

Diagram 4 6

Is 6 Nxf7 worth a try?

An interesting position!

43

S t a r ti n g O u t : A le k h i n e ' s D e fe n ce

6 Nxf7 K x £7 7 Q h 5+ K e 6 8 c4 The main alternative is 8 g3, a move which Tal spent a fair time ana­

lysing against Larsen, bu t sadly not until after he had played 6 Bc4 . The m ain line then is considered to be 8 ... b5 (to prevent c2-c4) 9 a4 (9 b3 b4 seems to be okay) 9 ... c6, but White has a powerful idea here in 1 0 Bh3+ (and not 1 0 axb5/Nc3 when Black can create a bolthole for the king on f7 by 1 0 . . . g6) 1 0 . . . Kd6 11 Nc3. The point of this is that 1 1 axb5 c xb 5 1 2 Nc3 allows 1 2 . . . N 7f6, since Black's king has the c6square, while now 1 l . . . N7f6 loses at once to 12 Bf4+ Nxf4 13 Qe5 mate. More subtly, 1 l . . . b4 12 Bf4+ Nxf4 13 Ne4+ Kc7 14 Qa5+ Nb6 1 5 Qe5+ elegantly forces Black to give u p the queen b y 1 5. . . Qd6, since 15 . . . Kb7 16 Nc5 is mate. Black needs a different eleventh move, for example 1 1 . . . e6 1 2 ax b5 N7f6 1 3 Qe5+ Ke7 may be possible. 8 . . . N5f6 9 d 5+ Kd6 1 0 Qf7 1 0 Bf4+? e 5 1 1 c5+ K e 7 1 2 Bg5 Qe8 leads now here. White c a n win his piece back with 1 0 c5+ Nxc5 1 1 Bf4+ Kd7 12 Bb5+ c6 1 3 dxc6 bxc6 1 4 Qxc5 s ince 1 4 . . . axb5 1 5 Qxb5+ Ke6 1 6 Qc6+ wins the rook, but un­ fairly enough 14 . . . Qb6! forces the exch a n g e of queens and gives Black a very reasonable position. He will trade the light-squared bishops with . . . Ba6 and anchor a knight on d5. 10 Qf5 Nc5 1 1 Bf4+ e5 12 Bxe5+ Ke7 (or 1 2 dxe6+ Kc6), 10 Qh3 Nc5 1 1 Q a 3 e5 12 Be3 b 6 13 b4 Ncd7 14 c5+ Ke7 15 Nc3 bxc5 1 6 bxc5 Kf7 1 7 Bc4 Nb6! and 10 Qe2 Nb8 1 1 Bf4+ Kd7 have all been tried without appearing to give ade­ quate compensation. 10 ... Ne5 1 1 Bf4 ( D iag ram 46) 1 l...c5

The only defence to 1 2 c5+. 12 Nd2

12 dxc6? Qa5+ 13 Nc3 Kxc6 w ins for Black at once as White's queen is trapped, but the major alternative is 12 Nc3. This forces 12 ... a6, and now refuted tries are 13 0 - 0 - 0 (with the idea of Re 1) 1 3 . . . g6! (threaten­ ing . . . Bh6!) 14 Bxe5+ ( 1 4 Re 1 Bh6!) 14 ... Kxe5 1 5 d6 Bh6+ 16 Kc2 Qe8 1 7 Rd5+ Nxd5 1 8 Qxd5+ Kf6 1 9 Ne4+ Kg7, and 1 3 b4 Qb6 14 bxc5+ (or 1 4 0 - 0 - 0 Qxb4 - 14 ... cxb4 is supposed to be a draw - 1 5 Kc2 Bf5+ 1 6 Bd3 Qxc4) 1 4 . . . Qxc5 15 Rd1 Qa3! 16 Rd3 Qc l + ! . For this reason 1 3 Rd l has been suggested t o improve o n the first try, a n d 1 4 Rc 1 ( 1 4 . . . g5 1 5 Bg3 h5!) o n the second, although i n neither case has White carried much conviction. 1 2 ... g 5

White was threatening 1 3 Nf3, although another p o i n t of h i s knight move i s revealed after 12 . . . Qe8?? 13 Ne4+! etc. 1 3 Bg3 g4

13 . . . Bh6 14 Nf3 Rf8 1 5 Nxe5 enables White to win the queen back and stay a pawn ahead, even a fter 15 . . . Qa5+ 16 Kd1 Rxf7 1 7 Nc6+.

14 0-0-0 ( D iagram 4 7 )

44

M a i n L i n e w i t h 4 . . .d x e 5

Diagram 47

Diagram 48

How does Black deal with the pin?

The bizarre events continue

Black c a n n ot defend e5, but he has a resource . . . 1 4 ...B h 6 14 Rg8 1 5 Rel R g 7 i s much t h e sa me, bu t a little worse. 1 5 . . .Rg5 ? 1 6 B d 3 B h 6 1 7 K c 2 i s m uch worse. . . .

1 5 R e i Rf8 1 6 Bxe5+ K d 7 1 7 Qe6+ Ke8 1 8 Qd6 (Diagram 4 8 )

A m ildly fortu nate resource for White i n his tu r n, forcing a level e nd­ ing. 18 Qxf6 R x f 6 19 B x f 6 Kf7 i s u nsou n d . 1 8 ... exd6 1 9 Bxf6+ K d 7 2 0 B x d 8 Kxd 8 2 1 Bd3 R x f2 2 2 R e 2 Rxe2 23 Bxe2 B d 7 2 4 Bd3 % - %

Statistics Black has scored 5 0% a fter 6 Nxf7.

Theoretica l ? I ntensely. Black really has t o k now t h e variations given here after the sacrifice a n d, fra n kly, i t's wise to h a ve done a bit more analysis of one's own, as a book such as this c a n provide no more than a gallop through some of the l i n es.

Theo retical Status 1 0 Q £'7 w a s the m ai n line for m a ny years, b u t it's looking a s i f White is the one trying to force equality. 8 g3 is far more worrying. Game 9 0 Piza • Mi roshnichenko Komercni Banka Open 1 996

1 e4 Nf6 2 e 5 Nd5 3 d4 d6 4 N£3 dxe5 5 Nxe5 N d 7 6 N£3

6 Bc4 is best answered by van der Wiel's logical 6 ... Nxe5 7 dxe5 c6.

45

S t a r t i n g O u t : A l e k h i n e ' s D e fe n c e

White really needs t o p l a y 8 Q f3 t o preven t Black reaching easy equality after 8 0-0 B f5 (or 8 Nc3 Be6), and now Black has 8 . . . Qc 7 ! , when n either 9 B x d 5 Qxe5+ 10 B e 4 f5 nor 9 0 - 0 Qxe5 1 0 N c 3 B e 6 i s a n y good , but defending t h e e·pawn with 9 B f4 r u n s into 9 . g5 1 0 Bg3 h 5 , when van der Wiel's suggestion w a s 1 1 Nc3 Be6 1 2 h 4. Here, however, Black has the very strong 1 2 . . . Qa5, when 1 3 hxg5 Nxc3 1 4 Bxe6 Nd5+ wins material, so White i s reduced to something miser­ able like 13 Bb3 g4 1 4 Qd3 Nxc3 1 5 Qxc3 Qxc3+ 1 6 bxc3 Bf5 . Maybe Wh ite ought to play a differen t eleventh move, but he won't stand bet· ter, e.g. 11 h 4 Bg4 . Meanwh ile 6 N x d 7 Bxd7 has no pretensions, for exa mple 7 Bc4 Bf5 i s equal . ..

6 ... c 6 (Diagram 4 9)

Diagram

49

A promising continuation?

D ia g ra m so Black is ready to contest the centre

6 . . . e6 a n d . . . c7-c5 has been much more popular, b u t the text has done better and - i n my opinion - has a lot to com mend it. 7 g3

After 7 c4 Nc7 8 Nc3 g6 9 Be2 Bg7 10 0-0 0-0 we would reach th e old mai n l i n e of the Miles variation (Game 4 , note to Black's 6th move) . 7 . g6 8 Bg2 B g 7 9 0-0 0-0 10 Rei ..

The natu ral 1 0 c4 Nc7 1 1 Nc3 is i nconveniently met by 1 1 . .. Nb6, for example 12 Qe2 B e6 13 b3 B g4 followed by . . . Ne6. 1 0 ... c 5 (Diagram 50) 11 d xc5

White has left it too late to drive the knight away from the centre, for example 1 1 c4 N5b6 12 b 3 cxd4 1 3 Nxd4 Nc5 1 4 Bb2 Qd7 followed by . . . Rd8 etc. 1 1. ..Nxc5 12 Nbd2

Not ideal, but by now 12 c 4 Nb4 i s out of th e question, so the knight cannot b e developed on its natural square. Perha ps 12 Qe2 was best. 12 N b 4 13 a 3 N c 6 1 4 Ng5 •..

46

Main Line with 4

••.

dxe 5

A bit of a desperate m easure, b u t . . . Bf5 was a difficult threat to m eet. 14 ... Bf5 1 5 Nge4 Nxe4 1 6 Nxe4 Q x d 1 1 7 Rxd l Rad8 1 8 Rfl Nd4 19 c3 Nb3 20 Rbl Nxcl 2 1 Rbxcl b6

Black i s better; the rem a i nder of the g a m e is i nstruct ive, a lbeit out­ side our theme. 22 f4 e5 23 fxe 5 Bxe5 24 Rcdl Kg7 25 B f3 Rfe8 26 Rxd8 Rxd8 2 7 Kg2 h 6 2 8 R e i Bb8 2 9 R e 2 B e 6 3 0 R d 2 Rxd2 3 1 Nxd2

I n principle this i s a n achievement for White, b u t Black's advantage is still considerabl e. 3 1. . . f5 3 2 Bc6 Kf6 33 N f3 g 5 3 4 h 3 Bc8 3 5 BaS h5 36 Nd4 a6 3 7 N c 6 Bc7 3 8 Nb4 Ke5 3 9 B f3 h 4 4 0 gxh4 gxh4 4 1 Nd3+ Ke6 4 2 c 4 B d 8 4 3 c 5 b xc5 4 4 Nxc5+ K e 5 4 5 Be2 a 5 4 6 b 4 Be7!

A n excellent move. Possibly White had been banking o n 4 6 . . . axb4? 47 a xb4 Kd4 4 8 Ba6 with a draw. 47 Nd3+

Now 4 7 B a 6 Bxa6 4 8 Nxa6 a4 49 Kf3 Bd6 traps t he k night and wins. 4 7 ... Kd4 48 bxa5 Ke3 49 Kfl f4 50 N� f3 5 1 B b 5 Bxa3 52 Kgl B c 5 5 3 Ng4+ Kf4+ 54 Kh 2 B b 7 5 5 a6 BaS

A c lassical demonstration of the abili ty of the b i shop to play on both wings - the a-pawn i s a l most irreleva nt. Black i s w i n n i n g by now. 5 6 Bd3 Kg5 57 Bfl Ba7 58 Bd3 � 59 Nx� Bx� 60 Bfl Bc6 6 1 Bg2

White can not keep h i s bishop o n the f l - a 6 diagonal: B l ack plays ... Ba7, b ri ngs the king u p to f2 and m ates w ith ... Bb8. 6 1 ...Bb5 62 Bb7 Bfl 63 Bc8 K f4 64 Bb7 Ba7 6 5 Bc8 Kf3 6 6 B b 7+ K� 6 7 Bc6 Bxa6 68 Bg2 Bb8+ 6 9 K h l Bc8

Zugzwang again! 70 Bc6 Bxh3 0-1

A fine technical grin d o f a type to gladden the heart of Tony Miles him self. ,

Stat istics White has managed 66% after 6 Nf3, but only 50% after 6 . . .c6, a lthough I ha ven't accounted for the transpositions to the Miles variation.

Theoreti c al? Not really. It's worth trying t o remember v a n d e r Wiel's cute sequence a fter 6 B c4 .

Theoretic a l Status N o t many g a m es, b u t I doubt White has much better than ing to the M i les system .

transpos­

47

Chapter Two

M ain Line with 4 . . . B g 4

• • • •

Introduction Black Plays 5 Black Plays 5

. . .

. . .

e6 c6

Illustrative Games

M a i n Li n e w i t h 4 . . . B g 4

Introduction 1

e 4 N f6 2 e 5 N d 5 3 d4 d6 4 Nf3 Bg4 (Diagram 1 )

D ia gram 1

Diagram 2

An i m mediate pin

The standard 5 e6 ...

This, a very natural way to develop, is Black's oldest move and was the most common choice u ntil the 1 97 0's, when the rise of 4 g6 and then the modern 4 . . dxe5 systems reduced its popularity. Its most outstanding devotee was the great Alekhine player Vladimir Bagirov. ...

.

5 Be2

Other possibilities a r e covered i n t h e notes to G a m e 1 0 but t h i s is far the most common and natu ral continuation.

Black plays 5 e6 . . .

1 e4 N f6 2 e5 Nd5 3 d4 d6 4 Nf3 Bg4 5 Be2 e6 ( D i agram 2 ) 6 c4 Nb6 7 Nc3 Be7 8 h 3

Believe i t o r not, this i s a very controversial point. Whether White should play this or not depends on what system he proposes to employ in the coming moves. If he is going to o p t for Bologan's idea set forth

in Game 1 2 , or for the 1 1 cxd5 idea in the notes to Game 1 0, then he needs to play h2-h3. I f he i ntends to play the main line of Game 1 0 then he would like to get by without nudging the h-pawn i f h e can, for the reasons explained in that game. On the other hand, if h2-h3 is not included, then Black can try the line seen in Game 1 1 . White there­ fore has to decide whether (a) Black is going to go for the main line anyway, whatever he does , i n which case he s hould avoid h2-h3, or (b) whether he would rather play the ma in l i n e even handicapped by having played h2-h3, o r play against the system seen in Game 1 1 . s 8

...

..

Bh5

. B xf3 i s not good: 9 Bxf3 N c 6 1 0 Bxc6 bxc6 leaves Black w i t h no

49

S t a rt i n g O u t : A l e k h i n e ' s Defe n c e

compensation fo r h i s paw n weaknesses. 10 exd6 cxd6 1 1 d 5 e x d 5 1 2 Bxd5 i s also quite strong. 9 0-0

The alternative is 9 exd6 cxd6 a n d now either Bologan's 1 0 d 5 (Game 1 2) or the older and still respectable 10 0-0. A variety o f exd6 systems is dealt with in Game 1 3 . 9 ... 0-0

A bad idea would be 9 . . . dxe5 1 0 Nxe5 Bxe2 1 1 Qxe2 Qxd4; White's initiative is too dangerous , e.g. 12 Rd1 Qc5 13 b4 Qxb4 1 4 Nb5. 10 Be3 d 5

The natu ral 1 0 . .. Nc6 i s weak - White replies 1 1 exd6 cxd6 1 2 d 5 a n d stands w e l l for t h e reason explai ned i n t h e notes t o Game 1 1 . 11

c5

Possible i s 1 1 cxd5 , w h ich is considered i n the notes to Game 1 0 . l l . . . Bx£3

Black plays this at this mom e n t p a r tly becau se the position is now go­ ing to become closed and h e n c e his k n ights will b e the equal of White's bishops, partly because with the centre fixed g 2 -g4 is becom­ ing more of an optio n for White, a n d partly b ec a us e he wants to b r i ng the kn ight to the active c4 -square if White recaptures with the bisho p. 12

gx£3!

This rath e r u n n atural move is credited by Alburt to Karpov, although in fact it seems to have been first played by Nona Gaprindas hvili, the perennial wome n's world champion of the sixties and seventies (and collected a qu estio n mark from comme ntators as perceptive a s Ea les and Williams, which shows just how u n natu ral it is). The older 1 2 Bxf3 Nc4 i s also perfectly good a nd is seen i n the notes to Game 1 0 . W i t h the te xt, though, W hite forces the b 6-knight b a ckwards and thus assures himself of a dom inating spatial adv a n tage . 1 2 . . . Nc8

(Diagram 3)

The best square, o n the way to the natural outpost o n f5 . This p os i ­ tion h a s been the start for many hundreds of contests. From Black's point of view it will suit players who are tempera mentally suited to prolonged trench warfare. His pieces are presently passive and disor­ ganised; he has no obvious pawn break and for many moves will need to co ncentrate on restraining White's possibilities and on getting his pieces to sen sible squares. There are compensations, though. White's weaknesses on the f-file are not a pretty sight, his queen's bishop is not a good piece at all, a n d f5 beckons a black k n ight. In addition, in the long term White's king's paw n cover is a litt l e disru pted, and when the position does open up this ca n often be a source of cou nter­ play. B lack has not scored terribly well on the whole but that is no cause for despair - maestros such a s Bagirov a n d Albu rt made big scores from this p os itio n .

50

Main L i n e w i t h 4. . . B g 4

Dia g ra m 3

Diag ra m 4

A 'start ing' position

An interesting alternative

Black plays 5 c6 . . .

1 e4 N£6 2 e5 Nd5 3 d4 d 6 4 Nf3 Bg4 5 Be2 c6 ( D i agram 4 ) With this move Black is ai ming for a sligh tly different set-up to 5 . . . e 6. His im mediate i de a is to take th e tension out o f t h e centre w i th the s eque n c e 6 . Bxf3 7 Bxf3 dxe5 8 dxe5. To do this, of course, he first ..

n eeds to cover th e b7-square. The other way o f doing this, 5 . . . Nc6, has the drawback that after 6 0-0 B xf3 7 Bxf3 dxe5 8 dxe5 e6 9 c4 Nb6 White ca n play 1 0 Bxc6+, rui n i n g Black's pawns. White, in h is turn, has two approaches against 5 . . . c6 - he can either p la y 6 0-0 and allow Black's idea (Game 14) or he can use one of two ideas designed to pre­ vent it, broadly Exchange-type systems (Game 1 5) or lunges with Ng5 (Ga me 1 6) .

Rather than a conventional discussion o f Black's i d e a s in the 6 0-0 variation, I think a 'How Good Is Your Chess' style feature i s called for. We have the usual deal - cover the page and try and guess White's next move - and you are sitting alongside England's talented young IM, Adam Hunt. Your opponent is the English international master Colin Crouch. The gam e was played in the Portsmouth Open, 2003.

6 0-0

No poin ts for th is. I did say this was the move. 6. . . Bxf3 7 Bxf3 1 point.

7 dxe5 8 dxe5 ...

1 8

point. ...

e6 9 Qe2

3 points. Lending extra support to the e -pawn and vacating the d -file for the fl-rook.

51

S t a rt i n g O u t : A le k h i n e ' s D e fe n c e

9 . . .N d 7 1 0 c 4

3 points. Driving Bl ack's knight t o a n out-of-play position o n b6. 10 ... Ne7 1 1 b3

3 poi n t s . I n te n ding Ba3 or Bb2 to support the e-pawn. 1 1 . .. Ng6 1 2 Bb2

1 point only. I gave you a h int last move. 12 . . . Qc7 13 B h 5

5 points. O f course not 1 3 Re 1?, which loses the e-pawn t o 1 3 . . . B b 4 (a very common square for the bishop in this line , by the way) . But now White i s ready to meet 1 3 .. . Nxe5 with 14 f4 (- f5 ), catch ing the black king i n the centre. Now add up your points: 17 points - Very good. Tal ented IM, perh a ps even gra n d m a st e r . Anything less th a n that - well, never mind. At least you aren 't almost lost on move 13 with the white pieces, like Hunt is.

We can all agree that White's sequence has been very natural, and a n u mber of grandmaster games have gone this way, but Wh ite might in fact be lost (and h a s been for a while, perhaps since 1 1 b3). Black can conti nue w ith 13 ... Ndxe5 14 f4 Nd7 15 f5 0 -0 -0! when 1 6 fxg6 hxg6 wins the piece back with i n terest because of the pin of the h 5 bishop against t h e h2-pawn, so Wh ite has to play 1 6 fxe6 fxe6 1 7 Qxe6, when 1 7. . . Bd6 with the idea of . . . Nf4 i s very strong for Bl ack. You may say that this is j ust a trap and proves nothing, and o f course you're right. But at least now w e know what Black wants - to attack th e e-pawn with all h is pi eces, and to castle queenside and perhaps look fo r his play on the kingside . And actually it's n ot j ust a trap - White's play has been anti­ posi t i o nal in various regards . Let's take another look: 9 Qe2

Not a good move. 9 Re 1 is better. The rook does not belong on d l . It h as nothing to do there except be exchanged for Black's rooks . One of the main advantages of havi ng more space is t h a t it tends to make your rooks better than your opponent's. Unless there is some concrete point to their being on an open file, they belong on files like the e-file here, where they can be moved forward and can then use their extra mobility along the open ranks the space advantage opens u p . Mean­ while a single bishop o n , say, c3 c a n keep even doubled rooks o n the d-file at bay. Once it is realized that the rook should be on e 1 , obvi­ ously it should go there at once. The e-pawn needs defending and the rook needs to be on e 1 . TIP: When planning your development the pieces which have no choice ( like the rook in this case) shou ld be developed first. The most flexible pieces ( like the queen and queen's bishop here) should be left unti l later, when their best position may become apparent.

52

M a i n L. i n e w i t h 4

•..

Bg4

The queen may be good o n e 2 , £'3 , g4 or h5. The q u ee n' s bishop may go to g5, f4 , e3 or b 2 , or it m a y even be good where it is. The rook and queen's knight are the pieces whose best squares a re already clear. 9 .. . Nd7

10

c4

Another mistake. Black can never prevent c2-c4. The longer he has to plan his game not knowing when or whether his knight is to be dis­ lo d ge d , the harder it will be for him. There are two positio n a l princi­ ples a t work here. One, it is almost always to the advantage of the side which can release a particular tension to maintain that tension as long as p o ss ibl e unless some concrete a d va n ta ge can be ga i ne d . Second, it is notoriously hard to plan o ne ' s game when one does not know where one's knights will be i n five or ten moves' time. Black must be left i n that u n h appy state as long as possible. 10 . . . Ne7 1 1 b3

i s a bad move largely because it misses the threat. It is h i gh time play something like 1 1 Bg4, so that at least White can play 1 2 f 4 a n d not lose the king's pawn. Thi s

to

Illustrative Games Game 10 0 Sanakoev • G.David Correspondence 1 988-92 1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 N£3 Bg4 5 Be2 e6 6 c4 Nb6 7 Nc3 Be7 8 0 - 0 0-0 9 h3 B h 5 10 B e 3 d 5 ( D i ag ram 5)

11

D iagram 5

D iagram 6

Standard

Structures

c5

1 1 cxd5 is an alternative. Since 1 1 . . . Nxd5 1 2 Qb3 Nb6 1 3 d5 is irritat· ing, Black usually plays 1 l . . . exd5, and now White usually goes 12 g4 Bg6 1 3 Ne1 these days (the older 1 2 N e 1 Bxe2 1 3 Qxe2 is not so test-

53

Starti n g O u t : A l e k h i n e ' s D e fe n ce

i n g) . Black cannot stand by on the k i n gs id e ; he ought to play 1 3 . . . f6 14 f4, and now either the solid 14 .. .f5 (Black should not pl ay this be­ fore a pawn has arrived on f4 because White will establish a strong knight there, although, to b e honest, if Black is goi ng to push the f­ pawn then he m ight have considered a different move at move 1 3, for exam ple 1 3 . . . Nc6, since White's plans surely include f2-f4 a nyway) , or the extremely ambitious 14 .. . fxe5 15 dxe5 of King-Baburin, 4 NCL 2000, in w hich Black lost after huge com p l ications, but not because of the opening. ll ... Bxf3 1 2 gxf3 Also very possible is 1 2 Bxf3 Nc4 13 Bf4 Nc6 14 b3 N4a5. Black can then proceed with . . . Qd7, . . . f7-f6, . . . Bxf6, . . . Ne7 and . . . Nac6 with a solid position, while . . . b7 -b6 at some point may also be useful. 12 ... N c8 ( Diagram 6 ) It i s worth considering this position for a moment. White's structure

simply demands that Black send a knight to f5, and indeed his most basic manoeuvre i s . . . Bh4 and . . . N8e 7 . I t will usually be n ecessary to play . . . g7-g6. Meanwhile ... Kh8 may well be necessary a s a reply to Rg 1 (otherwise White may have the tactical trick Qh5, attacking the bishop on h 4 a nd so forc ing the knight on e7 to move, either to f5, which will lose a pawn, or back to c8, when f 4-f5 will be possible and m a y well be crushing). The other knight will evidently go to c6. An­ other tactical trick is . . . Bxc5, dxc5 d4 regaining the piece. This needs c a r e fu l ti m ing; it may lead to the premature opening of the po s i t i on and Black getting crushed down the central files, or it may be good. I n t h e longer term, Black w i l l need either t o p u t h i s f-pawn on f6 o r f5. In the latter case he will tend to follow up with a thrust of the g­ pawn. To make this work he may well need to play . . . Ne7-g8-h6-f7 and then . . .Rg8 and perhaps even ... Rg8-g7 and ... RagS (or ... Qg8). Of course it's a s h a m e to dissolve the weakness on f4 , but if Black can ac· tivate his p ieces White still has plenty of pawn weaknesses left, and this is really t h e o n l y way of d oi n g that. Black will not normally want to play actively o n the queenside, but ... a 7 -a6 is usually a good idea. White should not be allowed to establish pawns o n a5, b 5 a n d c5, but exchanging a pair o f pawns and even a pair of rooks on the a -file is not something Black usually objects to. As for White, unless he manages to strike a t once on the kingside he will normally play on the queenside, manoeuvring to try and keep Black's ideas from working before usi n g his space advantage. Perha p s the main thing to a void is losing patience and tr adi n g the king's bishop for a knight on f5 - unless it is either forced (a bad sign) o r connected w i t h a n i m m ediate p l a n t h i s is seldom a good idea. Often White brings his knight round via e2 and gl to f3; as a rule Black does not mind this capturing a bishop on h4. This general talk is all very fine and large, of course, but in the mean­ time Black is undeveloped and is i n a certain amount of danger of getting crushed, so he does need to b e quite careful over the next few moves.

54

M a i n l i n e w i t h 4. . . 8 g 4

One deceptive point i s that White i s much better off i f he has NOT play e d h 2 - h 3 . '!'his square can be conv en i e n t ly occupied by his major pieces. In general, for example, he can molest the h4-bishop by Qg4, which may prevent . . . N 8e 7 . I f h 3 is occupied, then Black can reply with ... h7-h5, and the queen cannot then maintain the attack on the bishop and N 8e7 -f5 can follow. I f White can play Qh3, though, then ... h 7 - h 5 merely creates a weakness and Black cannot bring much­ needed reserves to the kings id e Here is a sample of how White needs this square to pursue his attack ( D i agram 7) . . . .

.

D i a g ra m 8

D i agram 7 Gone wrong

for Bla c k

Watch the h3-square

In this position , fro m Kobalij a-Nalbandian, Moscow 1 999, Black has not handled things well, and went down in flames after 1 7 ... Na6 1 8 Bxf5, si nce 1 8 . . . gxf5 1 9 Bh6 Rg8 2 0 Rxg8+ Qxg8 2 1 Rg1 i s un thin k ­

able. I f th e p a w n had been on h3, however, h e could have fought with 1 7 .. .f6! . Now if 1 8 e6 Re8 Black is in the game. If Nalbandian had tried that move i n the game, though, 19 Qh3! would have been crush· ing, since 19 ... B g 5 20 f4 drops a piece , and 19 g5 20 Qxf5 is horrible. . . .

Here, in Ulybin-Morovic Ferna ndez, Las Palmas 1 99 2 (Diagram 8), again White has gone for it with the front f-pawn, but this time Black has defended himself a little better. White could have played 2 3 Bxf5, b u t a fter 2 3 . . . N e 7 2 4 B d 3 b6 there i s n o particular guarantee that his attack is goin g to succeed. Instead, ignoring the missin g pawn, Ulibin brought everyone to the party with 23 Ne2 Ne7 2 4 N f4 Rg8 25 Rg3 Qf8 26 Qh4 ReS 2 7 Rag l , a n d blasted through a fter 2 7 . . . h6 28 e 6 Kh7 29 exf7Qxf7 3 0 Rh3 1 -0 . I m pressive, but what if the pawn had been on h3? Then the final move of the game would not have been possible, and ind eed Black would possibly even have stood better. Further­ more, earlier on, 24 ... Ng8 would have won White's queen, so he would perhaps have had to settle for 23 Bx f5 after all - the difference be­ tween a n unclear game and Black being murdered i n his bed. I could give more examples but, actually, I thi n k w e need a sloga n :

55

Starti n g O u t : A l e k h i n e ' s D e fe n ce WAR N I N G : I f Black plays this variation when White has not thrown in h2-h3, he runs a grave risk of being mated.

Enough talk - let's see some moves. I'm going to assume that h2-h3 has bee n played. 13 f4 Nc6

Black prepares to meet f4-f5. The obvious method of doing this, 1 3 g6?, fails horribly to 14 f5! exf5 15 Nxd5 Qxd5 16 Bf3 and Bxb7. The text is the most popular but it is critical to my m ind that Black can prevent this kn ight being driven to a5 in a move or two's time (see the next note). If he can't, then instead perhaps he should play 13 . . . Bh4 (which has been tried) or indeed 13 .f5 (also played). . . .

..

14 b 4 ( D iagram 9)

Diagram 9

Diagram 1 0

Another white pawn advances

What should White play now?

1 4 Bd3 g6 1 5 f5 is another sharp possibility which can be met by

1 5 . . . exf5 16 Qf3 Bxc5 with complications, as i n the game Chibur­ danidze- Bagirov, M insk 1 98 3 . 1 4 . . .B h 4

Naturally 1 4 . . . N x b4 1 5 Rbl and Rxb7 is n o t w h a t Black wants, b u t 14 ... a 6 is controversia l. In principle Blac k would l i k e to play this. The problem some perceive is 15 f5 exf5 16 f4 . White is now threatening to win the d-pawn with 1 7 Bf3, and so the only move to defe n d it is the inglorious 16 ... Nb8 when, according to Alburt, White s imply wins with 1 7 Bf3 c6 18 Qb3 followed by Nxd5, obtaining two pawns and a crushing p a wn centre for the piece. Others - Bagirov, for example ­ clearly do not agree, a n d I believe they are right a n d that 1 4 . . . a 6 is best. Black can perfectly well play 1 5 f5 exf5 1 6 f4 B h 4 ! 1 7 Bf3 N6e7! (not 17 . . . N8e7, when 18 Qb3 wins the d - pawn after all). With the pawn on h2 White sim ply continues with Bg2, Qf3 and Qh3, when Black is sorely pressed to stay afloat at all, but with the pawn o n h3 the bishop is safe a n d Black can unravel gradually (the knight on c8

56

M a i n L i n e w i t h 4...B g 4

can b e brought t o life with . . . b 7 -b6 o r via e 7 once Bl ack has played . . . c 7 -c6 and ca n move the e7 knight) . White's idea would not have worked at a l l before b 2-b4 and . a 7-a6 were i nserted - Black could have met 1 4 f5 exf5 15 f4? with 15 . . . Bxc5 ! . 1 4 f5 has been pl ayed, but with the idea o f 1 4 exf5 1 5 Bf3. This, however, i s well met by 1 5 . . . Bg5 16 Nxd5 f4! , when either 17 Be l N8e7 18 N x e 7+ Qxe7 or 1 7 Bxf4 Bxf4 1 8 Nxf4 N8e 7 ! (and a load o f theory b ased around t h e game Aseev-Bagirov, Berlin 1 990) have been fou nd to b e fine for Black. . .

. ..

1 5 b5

This h a s to b e the critica l m ove in m y opi n ion. White h a s not done es­ peci ally w e l l after the al ternatives Rb l , a 2-a3 or Kh2, all of which lead to quieter play. Black s hould meet any o f those with ... a7-a6; once he has secured h i s k n ight on c6 h e is free to play th e useful m oves o n the kingside. 15 f5 i s also possib l e, when 15 . . . exf5 16 b 5 ! is still not good; either 1 5 ... N8e7 or perhaps 1 5 . . . a 6 is better. 1 5 ... Na5

This is not what Black wants to play, but the problem is tha t after 1 5 . . . N6e7 16 Bd3 g6 ( 16 . . . Ng6 1 7 f5 Nxe5 1 8 dxe5 d4 1 9 f6 is even worse, while 1 7 . . . exf5 18 Bxf5 N8e7 19 Bd3 Qd7 20 Qg4 must also be better for White) 1 7 Qg4 his b ishop is an noyingly en prise, and after 1 7 . . . N f5 1 8 Bxf5 exf5 1 9 Qf3 c6 2 0 bxc6 bxc6 2 1 Rfb l the position opens u p before he is ready to contest the open b-file. 16 Bd3 g 6 1 7 Qg4

White might a lso try 1 7 f5 exf5 18 Q f3 c6 1 9 Bh6 Re8 20 Qf4 w i th at­ tacking chances, as i n Ulybin-Fernandez, above. In that game the pawn was on h2 and White's attack prevailed, but the presen t situa­ tion would b e unclear. 17

...

Kh8

This was the recommendation of Bagirov himself, but in Macdonald­ Palatnik, Hastings 1 999, Black - doubtless warned by the presen t game (Sa n akoev's collection o f h i s games h a d j ust b e e n published) played 1 7 .. .f5 (Bagirov h a d played 1 7 ... h5) . This h a s a lot of merit i n m y opinion. If White is going t o meet . .. f' 7 -f5 with exf6, far better tha t Black h a s n o t played . .K h 8 a nd . . .R g 8 first, misplacing these two pieces. Tha t game continu ed 1 8 Qg2 N e 7 19 Kh2 b6 20 c6 (if White wants to win he needs a different move to this; there are plenty con ­ ceivable) 2 0 . . . a 6 2 1 a 4 Nc4, w h e n White settled for the d r a w with 22 N e2, allowing a complete block after 22 . . . N x e 3 2 3 fxe3 a 5 (some wood­ shifti ng fol lowed but already the draw is clear). .

1 8 Kh2 Rg8 19 N e 2

A typical ma noeu vre, planning N g l - f3, annoying the h 4 - bishop a nd eyeing g5 . 1 9 . . . c 6 20 a4 f5 ( D i agram 10)

This had been the plan mapped ou t by Bagirov, b u t a surprise awaits Black.

57

S t a rt i n g O u t : A le k h i n e ' s D e f e n c e

2 1 exf6!

A really fine move. I can't think of a nother game in this variation where White has gone in for exposing his weaknes ses on the f-file i n t h i s way. Sanakoev's explanation is d isarmi ngly simple. With s u c h a bishop as White's on e3 the game cannot be won. It must have diago­ nals opened for it and to do that Wh ite needs to play this move and then put his k night o n e5. 21 . .. Qxf6 22 Ng1 h5

Forced i n order t o save t h e bishop (Bl ack can not retreat the queen and bishop and allow h4- h5) . White's plan depends on this poi nt. 23 Qd1 Ne7 24 Nf3 Rad 8 25 Rbl Rg7 26 Qe2 Nc4 White's plan was fou nded on the poor placement of this knight, a n d sooner o r l ater Bl ack h a d to try t h i s t o get it i n to the game. T h e trou­ ble is that Wh ite can ignore it for the mo ment and continue to strengthen his position.

27 Rb4 Nf5 28 Rg1 Qe7 29 bxc6 bxc6 30 Bxc4 dxc4 3 1 Ne5 Qe8 32 Qf3 ReS

3 2 ... N xd4? loses to 3 3 Bxd4 Rxd4 34 Nxg6+ Rxg6 35 Qxh5 + Kg7 36 Rxg6+ Qxg6 3 7 Rb 7+ Kf6 3 8 Qe5 mate. 3 3 Rxc4 Bf6 3 4 Rb4 Bxe5

B lack can not tol erate th is knight, but now Wh ite's dark -squared bishop catches a glimpse of the promi sed land - f6. 35 fxe 5 Kh7 36 Qe4 Qd8?

Sanakoev points out that 36 . . . Nxe3, with a hard fight still ahead ­ however Wh ite recaptures - was essential . 36 Bg5 wasn't possible to avoid this as Black could reply 36 ... a5 3 7 Rc4 (to defend the c-paw n) 37 . . . Rb8, taking the b - file. 3 7 Bg5 ( D iagram 1 1 )

58

Diagram 1 1

Diagram 12

The bishop comes alive

Find the threat . . .

Main Line with 4. . .B g 4

3 7 . . .Q a 5 3 8 Rg b 1 Qa6 3 9 Kg2 R d 7 4 0 R 1 b 3 Rf8 4 1 Bf6 Ng7 4 2 Rb8! Rxb8 4 3 Rxb8 Qxa4 4 4 Q e 3 ! (Diagram 1 2 )

Th e cruel threat of 45 Qh6+ Kxh6 46 Rh8 m ate puts a n e n d t o a n y il­ lusions Black m ight have h a d; some m ight have thought i t better to a l low this than to d rag the game out a s Black does . 44 . . . g5 45 Bxg7 Rxg7 46 Qe4+ Rg6 47 Rb7+ Kh6 48 Qf3 g4 49

Qf4+ 1-0

A m a gnificent g a m e . M y notes were based on Sanakoev's in his excel­ lent Wo rld Champion at the Third Attempt. Wh ite's p l an h e re is not typical for this variation a nd depends u pon Black's knight bei n g driven t o t h e queenside, b u t t h e game was t o o good to miss, a n d the specific opening sequ ence which a l lowed this i s also im portant. This has been a very brief treatment of a com plex variation, and if th e reader wants to learn more, then playing th rough the games of Bagi­ ro v and/or Alburt from the position after move 1 2 is recom me n ded.

Statistics Black has scored 42% after 1 2 . . . N c8, a n d t h i s drops t o 3 1% if h2-h3 i s not inclu ded. O n the oth er h and Black d o e s a l o t b ette r if h e can avoid h av i n g his k night exiled early on to a5, h ence th e im portance o f the va riation i n th e note to move 1 4.

Theoretica l ? Not terribly, although i t is wise to know a bit about the tricks con ­ nected with f 5 , b5 , and such like round about moves 1 3 - 15.

Theoretical Statu s H a rd to say. Most m a sters think White is a little better with his extra sp ace, but a plan or sequ ence which pins down that edge is not so easy to pinpoint. I wou ld have thought Black would do well to find a move order which avoids the exact position after 15 b5 , but Baburin, for example, seems h appy enough to allow it. Game 1 1

0 Bo rriss • Baburin Berlin 1 992

1 e4 N ffi 2 e5 Nd5 3 d4 d6 4 Nf3 Bg4 5 Be2 e6 6 c4 Nb6 7 Nc3 Be7 8 0-0 0-0 9 Be3

All as i n previous gam es, but now Babu rin deviates. 9. . . Nc6 (Diagram 1 3 )

Black i s n o w threa tening 1 0 Bxf3 1 1 Bxf3 Nxc4 a s well a s 1 0 . . . dxe5 , so White has to take steps . ...

1 0 exd6 cxd6 1 1 d5

1 1 b3 d5 is the note to Black's tenth move in Game 1 3 .

59

Starti n g O u t : A l e k h i n e ' s D e fe n c e

1 1 ...exd5

They used to play 1 l . . . B xf3 1 2 Bxf3 Ne5 13 dxe6 fxe 6 14 Bg4, but re­ sults were depressing and this position m u st be good for White.

Diagram 1 3

Diagram 1 4

Black holds back the d-pawn

Decision time for White

1 2 Nxd5

The alternative is 12 cxd5 Bxf3 13 Bxf3 Ne5 which is equal after 1 4 Be2 Nbc4 or 1 4 Bd4 Bf6 1 5 Be2 Nec4. 12

..•

Nxd5 1 3 Qxd5 B£6 ( D iagram 1 4 )

W e h a v e reached a potentially very i mportant position, although not one that has been much played la te ly . If Black can equalise here then he can also get the m a i n line in the more respectable version without h2-h3. It is important to note that i f h2-h3 Bh5 has been played then this position is significantly worse for Black since his bishop on h5 i s then en prise a n d he would have h a d to attend to t h a t rather than start his cou nterplay with ... Bf6. 14 Q d 2

A s ig n ifi c a n t choice. White's alternatives are 1 4 Rab 1 , 1 4 Rfd 1 a n d 1 4 Rad l . The last of these i s not effective: 1 4 Rad1 Bxb2 15 Qb5 ( 1 5 Rdb1 Be6! e qualizes a t once) 1 5 ... Bf6 1 6 c 5 d 5 1 7 Qxb7 Rc8 w a s fine for Black in Liss- Schrentzel , Israel 1 999, which continued 18 Bf4 Rfe8 19 Ba6 Re4 20 Bg3 Rb4 2 1 Bb5, and now 2 1 . . . Na5 22 Qa6 Bxf3 23 gxf'J Rxc5 24 a4 Nc6 would have been very adequate for Black a s 25 Bxc6 Rb6 regains the piece. A straight exchange of b-pawns like this i s not l ikely to favour White, s i nce it isolates his c-pawn to balance Black's d - pawn.

14 Rab 1 allows either 1 4 . . . Be6 1 5 Qb5 d5 with complications, or the more sedate 14 . . . Qc8 (threatening . . . Bf5) followed by 15 . . . Rd8. 1 4 Rfd l is well met b y Hart's cunning 14 . . . Rc8, whose tactical justification is 1 5 Qxd6 Qxd6 16 Rxd6 Bxb2 17 Rb 1 Ba3, when 18 Rd3? fail s to 18 . . . Bf5. Black also prevents 1 5 Qd2 in view of 1 5 ... B xf3 16 Bxf3 Ne5

60

Main Line with 4...Bg4

with equality. Instead White does best t o continue with 1 5 R d 2 Qa5 1 6 Qxa 5 ( 16 Qxd6 is another a mbush - 16 ... Bxf3 1 7 Bxf3 Rcd8 18 Qf4 Be5 wins) 16 . . . Nxa5 1 7 Rc l Be6 1 8 b 3 d 5 with rough equality, e.g. 1 9 c 5 Nc6 2 0 Bg5 ( 2 0 Ng5? d4) 2 0 . . . Bxg5 2 1 Nxg5 Rcd8 - perhaps White is a little better here, b ut how significantly so remains to be see n . 1 4 ... Qa5 15 Qxa5

A novelty. After the usu a l 15 Rfd l the best move is 1 5 ... Rfd8! , intend· ing t o m eet 16 Qc2 with 16 ... d 5 . Pe r h a p s Wh ite's b est line is then 16 h3 Be6 17 Ng5, but Black has 1 7 . . . B x g 5 18 Bxg5 f6 19 Bf4 Qxd2 20 Rxd2 d 5 when h e is close to equality - although White has the two bishops, Black's pi e c e s and pawns coordinate wel l . 1 5 ...N x a 5 1 6 Rab 1 Rfe8

Not 16 ... Bf5? 17 Rbd l Bxb2?? 1 8 Rd5 and White wins a piece. 17 b3 Nc6 18 Rbd l Rad8

I don't quite k now why Baburin didn't play 18 .. . Bg5. After 19 h3 Bxe3 20 hxg4 Bc5, 1 9 Rxd6 Bxe3 2 0 fxe3 Rxe3 or 19 Bxg5 Rxe2 Black seems fine to me, a lthough developing the final piece with the text is also n atural. 19 h 3 Bf5 20 g4 B g6 2 1 Bd3

A thematic exchange to gain control of d5. 2 1 . .. Bxd3 22 Rxd3 Nb4 23 Rd2 Bc3 24 R d d l b6 25 Nd4

White is slightly better. He might also have tried 24 a3 Nc2 25 Be l , or here 2 5 Rd3 Nxe3 26 fxe3, when th e g a m e is to try and bring the knight to d5. The text aims for a tactical resolution, but Baburin is up to the task . 25 ...Nxa2 2 6 N b 5 Re7!

Natural ly Black prefers to lose th e w e a k d-pa w n rather than the a­ pawn. 27 Rxd6 Rxd6 28 Nxd6 R d 7 29 Nb5 Be5! 30 R e 1

3 0 f4 B b 8 3 1 Re 1 f6! i s t h e sa me. 30 ... f6 3 1 f4 B b 8 3 2 Bxb6

This looks strong, b ut Baburin has it u nder control. 3 2 . . . Kf7 3 3 Be3 a6 The kn ight h a s n o good square - 34 Na3 Rd3, or 34 Nd4 Nb4 (or

34 . . . Bxf4) 35 Re2 (a nd not 35 Rdl? Nc2 36 Kr.2 Ba 7 37 Nxc2 Rxd l 38 Bxa7 Rd2+) 35 . . . Bxf4.

34 R e 2 axb5 35 Rxa2 Rd3 'h- 112 White m ight have considered playing o n with 36 Kf2 Rxb3 37 c5 Ke6 but the position is becoming do u bl e - e d g e d .

Stati stics Absolutely grim. From a s m a l l sample Black has scored 2 8% .

61

S t a r t i n g O u t : A l e k h i n e ' s Defence

Theoretical? Fairly. Black needs to p la y pretty accurately round a b out m o v e 1 4 - 1 5 .

Theoretica l Status Q u i t e good, considering.

White has not identified a me t h o d to obtain is clear that Black does have to play significantly more accurately th a n White for a few moves in the teens before he ca n equalise. For this reason, and because this variation does have the po t e n ti a l to make Black's life in t h e main line a fair bit e as i e r I have given quite a lot of detail round about move 14. much of

an e dge

.

Nonetheless it

,

Game 1 2 D Bologan • S.Agdestein Yerevan 1 996

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 B g 4 5 B e 2 e6 6 c4 Nb6 7 Nc3 Be7 8 h 3 Bh5 We've seen all this before, but now White deviates. 9 exd6 cxd6

1 0 d 5 (D ia g ram 1 5)

Diagram 1 6

Diagram 1 5

White seeks early action A difficult move to u ndersta n d , a l t h o u gh

perfectly

illustrated

by the

A

consistent advance

White's ideas with it are

present game.

1 0 ... e 5 na t u ra l reply. If Black agrees to accept pawn weaknesses after dxe6 he m u s t expect to be worse. I n view of the next move one might get the idea of 10 Bxf3 1 1 Bxf3 e 5 , but White has 1 2 c5 dxc5 13 d6,

The

. . .

winning material.

1 1 g4 This i s the point of White's idea. Earlier Bologan had played

62

1 1 Be3

M ai n L i n e w i t h 4 . . . B g 4

Bxf3 1 2 Bxf3 N 8 d 7 1 3 Q e 2 R e S 1 4 b3 a n d w o n brilliantly, but Black should now play 14 . . . 0-0, after which 1 5 g4 Bg5 1 6 h 4 Bxe3 is sup­ posed to be a touch better for White, but looks quite p l ay a bl e to me. It's worth m entioning t h e clever 1 1 c5?! dxc5 1 2 g4 Bg6 13 Bb5+ N8d7 1 4 d 6 Bh4 15 Bxd7+ Nxd7 1 6 Qa4. White is threatening to win a piece with g4 -g5, but the coo l 16 . . . Q f6 defends. The id e a of 1 1 g4 - of White's whole plan, in fac t - is to ensure co m p l e t e control of the e 4square. To achieve th is it i s necessary to prevent ... f7-f5, and also use­ ful to cut out the possib ility of Black's trading his b a d bishop with ... Bxf3 a n d . . . Bg5. White is committing himself to playing with his king either i n the centre or i n the open spaces on t he queenside, but he relies upon h is space advantage a n d the absence o f pawn breaks to protect him. It is also possible to play more quietly with , for example, 1 1 0-0, but the text is the most com mon as well a s the m os t thematic. 1 l . . .Bg6 1 2 h4 (Diagram 1 6)

An essential follow-up. White m us t be able to

answer . . . h5

with g4-g5.

1 2 ... h 6

I n my view this move i s taki n g a grave strategic risk. The contours of a good knight against a bad bishop s c e n a rio can already be perceived - White ha s won various games in this line by e xch a n g i n g bishops af· ter Bd3, t ra d in g a pair of k ni gh t s , capturing a knight on c5 wi t h his q u e en ' s bishop and winni ng with a knight on e4 a g a i ns t the poor b i s ho p on e 7 - see Guseinov-Aliev, B a k u 200 1 for a particularly ago­ n i z i n g example. If Black o p ts for the a l te r n a ti v e 1 2 . . . h 5 this cannot happen - in such an ending he will alw ays a t least have the h 4 - p a w n to a i m a t and the possibility of breaking with . . . f7-f6. Th e crucial posi­ tions arise a fter something such a s 12 . . . h5 1 3 g5 N8d7 14 Be3 Qc 8 1 5 b 3 N c 5 1 6 R g 1 Qf5, o r 1 4 Bd3 Q c 8 1 5 b 3 Nc5 1 6 Bx g 6 fxg6 1 7 B e 3 0 0 . . . theory h a s y e t t o pronounce finally a bout these. 13 Bd3

13 h5 Bh7 14 Bd3 is a mistake - Black has 14 . . . Bxd3 15 Qxd3 Qc8 (forking a couple o f pawns) 16 b3 Qxg4 17 Rg1 Qxh5 18 Rxg7. If Black had played this same way i n the game ... Qh5 would not have captured a pawn and this final position, with a pawn still a t h4, would have fa­ voured White. 13 ... Qc8

Black didn't have to agree to his pawn structure being damaged like this, o f course, but he hopes for f-file activity. The alternative 13 ... Bxd3 14 Qxd 3 N8d7 1 5 Be3 Qc7 16 b3 Nc5 is also better for White. 14 Bxg6 fxg 6 1 5 Nd2

Defending both pawns by w a y o f the trick 15 ... Nxc4? 16 Qa4+. 15 ...0 - 0 1 6 b3 Rf4

T h is looks scary, especially in view of 17 f3 Qd8 (although this is by no means tragic for White), b u t Bologan now shows the real power of h is idea .

63

S t a rt i n g O ut : A l e k h i n e ' s D e f e n c e

1 7 g5 ! hxg5?!

It was surely better to avoid 1 7 . . . Qf5.

this

and play either 17 ... h5 or perhaps

1 8 hxg5 Qf5

1 8 Bxg5? 19 Nde4 w i n s for White, but Black must w i n the g5-pawn i n the long ru n . The proble m is that this exposes his king and does nothing to combat the monster k night on e4 . . . .

1 9 Q e 2 Na6 20 Nde4 (Diagram 1 7)

Diagram 1 7

Diag ram 1 8

The e4-square i s crucial

A d ecisive knight return

20 ... Rf3 2 1 B e 3 Simply

developing. Black is positionally lost.

2 1 ...Nd7 22 0-0-0 Ndc5 23 Rh4 R h 3

Of course i f 23 . . . N xe4 24 Nxe4 Nc5 White must continue w i th 2 5 Bxc5! dxc5 2 6 Rg 1 with a horrid position for Black. 24 Ng3 Qd7 25 Rxh3 Qxh3 2 6 Qc2 Kf7 27 a 3

White calmly p repares t o d rive away t h e black knights a n d re-enter the e4-square. 2 7 ... Qg4 28 b4 Bxg5 2 9 Bxg5

Not, of course, 29 bxc5 B xe3+ 30 fxe3 Qxg3 3 1 cxd6 Qxe3+. White does not need these com plications even if they are good for h i m. 29 ... Qxg5+ 3 0 Kb2 Nd7 31 Nge4 (Diagram 18) 3 1 . .. Qh4

3 1 . . .Qe7 32 Nb5 is no better, and 32 c 5 is a l s o very strong. 32 Nxd6+ Kg8 3 3 f3

Simply preparing Qxg6. 33 ... Qf6 34 c5 Qx£3 35 Qxg6 Q�+ 3 6 Kb1 N f8 37 Qh5 Qg2 38 Nce4 Ng6 3 9 Ng5 Q g 3 4 0 Q h 7+ Kf8 41 Ne6+ 1-0

4 1 . . . Ke7 42 Q x g 7 is mate. A striking game. White's guiding idea

64

Main Line with 4... B g 4

shines through the play, b u t the c alculation involved i n m aking i t look s o s i m pl e against a player like Agdestein is a lways greater than i t seems.

Statistics White has scored 70% after 1 0 d 5 , although there haven't been that m a ny games yet.

Theoretica l ? Not i m m ensely so.

Theoretica l Status This line is a problem for B l a c k j u s t now. T i m e will tell how t h e tions a fter 1 2 . . . h 5 work out.

posi­

Game 13 0 Skaslien • Ca ranzano Correspondence 1 988

1 e4 Nf6 2 e 5 N d 5 3 d4 d6 4 Nf3 Bg4 5 Be2

It is possible to play 5 c4 Nb6 6 exd6 a l ready, a l though after 6 cxd6 the gam e will most likely transpose to the present game. 5 c4 N b6 6 d 5 !? is a lso a move, when Black's best reply is 6 . . . e6, and a fter 7 exd6 for my money Black should brave the s light com plica tions which arise after 7 . . . Bxd6 8 Qd4 Bxf3 9 Qxg7 Rf8 10 gxf3, alth ough the solid 7 . . . Qxd6 i s a lso possible . Finally, 5 h 3 Bxf3 6 Qxf3 dxe5 7 dxe5 e6 i s a lso m e t with occasionally, when Black should h ave a n easy gam e with . . . Nd7, . . . c7-c6, . . . Qc7 a n d perh aps . .. h 7 - h 5 (after Qg3), roughly as i n Game 10, although White's pieces are less well placed h ere. . . .

5 ... e6 6 c4 Nb6 7 0-0 B e 7 8 exd6 cxd6 (Diagram 1 9 )

Diagram 1 9

D i a gram 20

A release of central tension

The knight continues to roam

65

S t a rt i n g O u t : A l e k h i n e ' s Defe n c e

This i s a slightly old-fashioned system , b u t still popular i n a m ateu r chess. It rather resembles the Exch ange Va riation (see G a m e 2 1 ). 9 Nc3

White h a s two oth er ways t o play at this point. The first is 9 b3 (in­ tended both to guard against . . . Bxf3, Bxf3 Nxc4, although this is not n ecessary yet, and also to rule out the ... d6-d5 , c4-c5 Bxf3 , Bxf3 Nc4 sequence) . Black can reply with 9 . . . 0-0 1 0 Nc3 Nc6 11 Be3 d5 12 c5 Nc8 when play will d e ve l o p along lines very si mila r to G a me 2 1 . I f h2h 3 Bh5 h a s been inserted then Black does w ell to m e et c4-c5 with . . . Bxf3 a t once; otherwise g2 - g4 might prove an noying. He ca n con­ tin ue with . . . Bf6 and . . . N8e 7-f5 . Th e seco nd, similar, idea is 9 Nbd2 Nc6 1 0 b 3 0 - 0 1 1 Bb2 . This is a lso designed to take the sting out of Black's . . . d6-d5 pl an , this t i m e by e n ­ suring that Wh ite can recapture w ith a k n ight o n f3, keeping his bishop on e2 to a ssist on the queen side, and also ensuring that d4 is stoutly d efended. It i s certainly possible to play 1 l . . . d 5 a nyway, but usually Black has preferred to wait a little with 1 1 . . . Bf5 12 a3 a5 1 3 Re 1 (White i s a little short o f a good waiting move) and only now 1 3 . . . d5 14 c5 N d 7 with the idea of a quick . . . e6-e5 (White is not so well placed to m eet this with h is k n ight on d2 in stead of c3). 9 ... 0-0 1 0 Be3

White can also try the quite cunning 10 Re l . This aims to save a tempo by ensuring that any . . . Nc4 does not hit a bishop on e3, while a lso being useful a fter 10 ... Nc6 1 1 d 5 exd5 1 2 N x d 5 Bxf3 13 Bxf3. Black's best reply is 1 O ... a 6, which is often a useful move in the 4 . . . Bg4/e6 line. This threatens 1 l . . . Bxf3 and . . . Nxc4 by gi vi ng the aS­ rook a fl ight s q u a re, and will a lso be useful after ... d6-d 5 , c4 -c 5 by holding back White's b-paw n . 1 0. . .d 5 1 0 . . . N c 6 1 1 d 5 leads t o Game 1 1 .

1 1 c5 Bxf3 1 2 Bxf3

12 gxf3 i s possible, but b y com parison with Ga me 10 White's pawn weaknesses appear in sharper relief, a n d Black has decent play after 12 . . . Nc8. 12 ... Nc4 1 3 Bf4 Nc6 1 4 b3 N4a5 ( D i agram 20) 1 5 Rb 1 Bf6 16 Be3 b 6 1 7 Na4 White might prefer 1 7 cxb6 here, when opinions differ about whether Black should recapture with q ueen or pawn. The for m e r keeps control of b5, slightly more active pieces and the hope of a m inority attack

with . . . a 5 - a 4 ; the l atte r controls c5. 1 7 . . . b5 1 8 N c 3 b 4

A very solid pl an, b u t i f Black w a n ts t o pla y for a win he probably needs to permit an exchange of h is b - pawn for White's c-pa w n . 19 Na4 Ne7

66

M a i n L i n e w i t h 4. . . B g 4

A thematic rearrangement o f t h e k nights. 20 g4

To prevent . . . Nf5. 20 ... g6 Addressing

the threatened g4-g5 .

2 1 Qd2 Nac6 (Diagram 2 1 )

Diagram 2 1

D iagram 22

T he knights finally regroup

The pressure mounts

22 Rbd l Qd7 2 3 Bg2 Rfd8 A move designed to threaten .. e6-e5 a n d so elicit Whi te's next, which is ba d news for the e3 -bishop. .

24 f4 Bg7 25 R f3 f6 26 Bh3 f5

Of course. The e-pawn's plight i s regrettable, b u t Black cannot permit f4-f5. 27 g5 Kf7 2 8 B f1 Rh8 2 9 h 4 h6 Again Bl a ck cannot

permit h4-h5 -h6 followed by slow play on

the

queenside.

30 Rg3 hxg5 3 1 hxg5 Rh4 3 2 Kf2 Rah8 3 3 Kf3 Qc8 3 4 Q £2 Rh2 3 5 Rg2 R8h3+ 3 6 K e 2 Q h 8 (D iagram 2 2) 3 7 R d 2 Q h 5+ 3 8 K e 1 R h 1

Black i s i n total control, but White just m a n ages to defend. 39 Rg3 R3h2 40 Rg2 Rh4 4 1 Rg1 R 1 h 2 4 2 Rg2 R2h3 43 Rg3 R h 1 44 R g 1 R 1 h 2 4 5 R g 2 R2h3 !4-'h

Black i s forced to concede that he cannot make progress. Not a n excit­ ing game, b u t an instructive and correct one fro m Black's viewpoint.

Statistics Encouraging. Black h a s scored 5 2% after 1 1 c 5 .

67

St art i n g O u t : A l e k h i n e ' s D e f e n c e

Theoreti c a l ? No. The ideas i n this ga m e a re what Black needs t o rem em ber.

Theoretical Status Sound. Game 14 0 Gallagher • C rouch

4NCL 2002

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 N£'3 Bg4 5 Be2 c6 ( Diagram 2 3)

D ia g ra m 23

Dia g ra m 24

A d ifferent approach

Flexible

6 0-0

used to be thought of as a mistake, giving Black w h a t he wants, but now people are not so sure.

This

6... Bxf3

Black needs to make this exchange before playing . . . dxe5 , since dxe5 7 Nxe5 Bxe2 8 Qxe2 Nd7 9 f4 is n ic e for White.

6

. . .

7 Bxf3 dxe5 8 dxe5 e6 9 R e i

White can d evelop i n many other ways, for example b2·b3 a n d Bb2, but Bl ack's immediate development scheme is going to be the same in any case. 9 Nd7 10 Nd2 (Diagram 24) ...

This set-up is the most popu lar and the most flexible for the reason I mentioned i n the in troduction: White's queen and queen's bishop are the pieces which have the most sensible options available to them , so White develops the others first. IO

. . .

Qc7

10 . .B e 7 .

68

11

Nc4 0-0 is a less a mbitious set-up, but perhaps safer.

Main Line with 4...B g 4

1 1 Nc4 N 7 b 6

'I'he poi nt o f White's p l a y i s t h a t t h e natural 1 1 . ..b5 ? l o s e s material to 1 2 Bxd5 cxd5 13 Nd6+ Bxd6 1 4 Qxd5 ! ( 1 4 ... exd5 15 exd6+ wins the queen back ) . Bl ac k migh t have tried to avoid this trick by developing the queen o n bS, but th e straightforward 1 2 Nd6+ Bxd6 1 3 exd6 Qxd 6 1 4 b 3 Qc5 1 5 c 4 i s a lso e x tre m el y dangerous. 12 Qe2

1 2 N x b6 axb6 is less go o d

.

12 ... Nxc4 1 3 Qxc4 0-0-0 14 a3 h6 1 5 b4 g5 16 g3 Bg7 1 7 Bb2 h5 (Diagram 2 5 )

Diag ra m 2 5

Diag ram 26

Which king i s t he safer?

A logical sidestep

1 8 Qe4 g4 19 Bg2 N e 7 20 a4

Black h a s played a l l t h e u s u a l moves, a n d is ready for 20 Ng6, in­ creasi n g the pressure on e5, b u t where h e's actually go i n g is not clear. . . .

20 ... Rd2 2 1 Bc3 Rd7 2 2 Rad1 Rxd1 23 R x d 1 Rd8 24 Re1 ( D i a­ gram 2 6)

White wants to exc hange exactly one pa i r of rooks so t h a t he can sub­ sequently b u m p Black's re maining rook with Bg5, when Black must aban don either the d-file by moving the rook sideways, or else the k i ngside pawns by movi ng it to d7. I f Black had two rooks h e could assign o ne to each task, while i f both rooks were exchanged Wh ite would lack the firepower to exploit Black's slightly wobbly king. 24 . . . N d 5 25 B d 2 Nb6 26 Bg5 R h 8 27 b5 cxb5 28 axb5 N d 7 29 Qa4 Nxe5 30 Bf4 Q c 4 31 Qxa7 Qxb5 32 Qa2 (Diagram 27)

White has systematically opened u p the enemy's defen ces for his bishop o n g2 and now the threa t o f Rb 1 forces evacuation . 3 2 ... Nf3+ 3 3 Bxf3 gxf3 34 Qa8+ Kd7 35 R d 1 + Ke7 36 Qa3+ (Dia­ gram 2 8) 36 . Kf6 3 7 Qxf3 Qc6 3 8 Qd3 ReS 39 h4 Bf8 40 Q h 7 Qxc2 ..

69

Start i n g O ut : A l e k h i n e ' s D e f e n c e

Diagram 27

Diagram 28

The bishop pair dominates

Another forcing retreat

In the long run Black's king's position must be fatal, but Crouch must have been in time trouble to allow mate in two in this way. 41 Bg5+ Ke5 42 f4 mate 1-0

Statistics Black has scored 47% after 8. e6, but this falls to 33% if White goes in for the line shown here, a n d there have been quite a few games. . .

Theoretica l ? Not especially. It's worth recalling that winning the king's pawn with ... b 7- b5 is often not a good idea, but apart from that Black's game is initially quite easy to play. The proble m tends to come with finding active play in the m iddlega m e .

Theoret i ca l Statu s Solid, but White does tend to retain a n a gg i n g edge at a higher level. However, White's game is not at all easy to handle when seeing Black's system for the first time, for example the great Botvin nik came up with the moderately ludicrous manoeuvre Bf3-g4-h 3 i n a gam e against Flohr at Nottingham 193 6 (I assume - charitably - that was n't preparation). Game 15 0 Ferguson • Neil Norwich 1 994 1 e4 Nf6 2 e 5 N d 5 3 d 4 d 6 4 Nf3 Bg4 5 Be2 c6 6 c 4 ( D i ag ram 2 9) 6 ... Nb6

6 ... Nc7 is also possible . The knight is quite well placed here if White goes in for exd6 exd6, with easy access to e6 after ... d6-d5, c4-c5

70

Main L i n e with 4. . .B g 4

(unless perha p s Wbite can arrange a n early d 4-d5 ), and s o the ma in l i n e is co nside red to be 7 0 - 0 Bxf3 8 exd6 Bxe2 9 dx c 7 Bxd 1 10 cxd8Q+ Kxd8 1 1 Rxd 1 with something between equality and an edge for White.

7

Diagram 29

Diagram 30

White gains an early space advantage

An unconventional entry

Nbd2

7 N g5 Bxe2 (this s eems neces sary a s 7 . . .Bf5 8 e 6 fxe6 9 g4 Bg6 1 0 Bd3 Bxd3 1 1 Qxd3 is considered strong for White with the black knight exiled to b6) 8 Qxe2 dxe5 9 dxe5 e6 is similar to Game 1 4, although White is supposed to have benefited by retaining a knight instead of his king's bishop. Looking at what Gallagher did with his king's bishop, though, I'm not so sure about that. I t is also p os s ible to p l ay 7 exd6 now, but the a 2-a4 idea of this game is less strong while Black can still play ... N a6- b 4, so usually White will wait a move. 7 . . . N8d 7

7 . . . dxe5 8 Nxe5 Bf5 (or . . . Be6) can be played and h a s actually been more popular, a lthough Black has scored a miserable 28% and theory has a lways considered White better. On the other hand 7 . . . dxe5 8 Nxe 5 Bxe2 9 Qxe2 Qxd4 is a little brash: 10 Ndf3 Qc5 1 1 0-0 leaves White with too many attacking chances. 8

exd6

Black's pieces a re not a t a l l well placed for a n Exchange-type position a nd, furthermore, White has a specific idea in mind. 8 0-0 (or 8 h3) Bxf3 9 Nxf3 dxe5 10 dxe5 e6 produces play s imi l a r to G a me 1 4 White h a s a k night rather than bishop on f3 , which is nice for his e­ pawn, but equally he no longer h a s the option of sending the knight to c4. 8 Ng5 is another dangerous idea. 8 . .. Bf5, a iming for play simil a r to Game 16, s e ems to be in trouble owing to 9 g4 Bg6 1 0 e6 fxe6 11 Nxe6 Qc8 12 f4 with the simple intentio n of f4 -f5 (the k night on d7 hampers

71

S t a rt i n g O u t : A l e k h i n e ' s D e fe n c e

Black since it h a s t o move before he can captu re t h e pawn on e 6 after, s ay, 12 . . . Bf7 13 f5 Bxe6 1 4 fxe6, and it has no very convenient square - 1 4 . . . Nf6 15 g5) . On the other h a n d 8 ... Bxe2 9 e6! (a pulse-quickener if you've not seen it before) 9 . . . f6 10 Qxe2 fxg5 1 1 Ne4 ( 1 1 exd7+ is not so pressing as Bl ack can castle long and a i m for a quick . . . e7 -e5) 1 1 . .. Nf6 1 2 Nxg5 Nxc4! (trying to keep the piece is unwise b ecause White plays Nf7 and g2-g4 a n d develops a huge a ttack) 13 0-0 Qa5 ( 1 4 Nf7 threa tened) 14 Qxc4 h6 is reputedly okay fo r Black, although little tested. 8 ... exd6 9 0-0 Be7 10 Re1 0-0 1 1 a4 a5

Black does not care to ignore the th rea t of a4-a5, gaini ng vast a m ounts of s pace, although he might con sider 1 1 . . .d5 instead. Mter, for example, 12 c 5 Nc8 1 3 b4 Bf6 14 Ra3 it remains to be seen whether White's ea rly excu rsion on the q u e en s i d e has achieved a ny­ thing. Mter the text the outpos t on b4 is of little u s e to Black with his knights where they a re, and White's rook can be developed handily via a 3 ... 12 Ra3 (Diagram 30) 1 2 ... Re8

12 . . . B h 5 and 12 ... Qc 7 a re alternatives, but 12 . . . Bf6 is well met by 13 Ne4. 13 Re3

A relatively popu lar position. 1 3 ...Nf6

Black has also tried 13 .. . Nf8. The text is aimed at ru ling out alto­ gether the possibility of d4-d5. After 13 . . . Nf8 14 Qb3 the th reat of d4d5 is thought to fo rce the passive 14 . . . Be6, leaving White with a com­

fortable edge a fter 15 Bd3. 14 h 3 Bh5 1 5 b3 Qc7 A typical position in which White p e rh a ps has a p u l l, a l though in this game he now conceives a grandiose plan to dominate the k ingside which o u ght not to have succeeded. Instead he should play q uietly, beginning with 1 6 Bd3 or Qc2, for instance.

16 N h 2 . Bg6 1 7 Ndfl Bf8 18 Bf3 Nbd7 19 g4 Rxe3 20 Rxe3 ReS?

Black h a s reorganized sensibly up to now, b u t this is too meek . I n ­ stead 20 . . . d 5 (D iagram 3 1 ) was called for. T I P : The first reaction to a f lank attack should always be to look for a counterb low in the centre.

Now 2 1 g5 Ne4 is n o good, while after 2 1 c5 b6 White will soon regret over-co m m itting his knights to the kingside, b u t otherwise . . . B d 6-f4 is going to b e an noying. 2 1 Rxe8 Nxe8 22 Qe l f6

22 . . . Nef6?? 23 g5 Nh5 2 4 Qd 1 sees White win mate rial, b u t now . .. d6d5 is ruled out and White i s b ack on tra ck.

72

Main Line with 4...Bg4

Diagram 3 1 The central

counter-strike

Diagram 32 White's kni ghts have the

last dance

23 Bd2 b6 2 4 Be4

light-square strategy. Wh ite wants to trade d-pawn.

Classic

the

bishops

and push

24 Bf7 2 5 Q e 3 NbS ...

A feeble move. 25 d5 was essential, and works tactically a s 26 cxd5 cxd5 27 Bxd5? Bxd5 2 S QxeS B xb3 is poor for Wh ite. For this reason Wh ite's last move was po s s ibl y slightly inaccurate and needed to be replaced with someth ing l ike 25 Ne3. ...

26 Q£3 Q d 7 27 d5 c5

Now White has what he wants and stands con siderably better. Black cannot leave the do mina ting bishop on e4 where it is, but if he trades it h i s remai ning bishop is a sorry sight. 2S h4 Bg6 29 Bxg6 hxg6 30 Ng3 Q f7 31 Qe4 Nc7 32 f4

There was nothing wrong with an immediate 32 h5, but Ferguson prefers to ru le out even the possibility of .. . g6- g5 in reply. 32 ... Nd7 33 h 5 gxh5 34 gxh 5 NbS 35 Kg2

Always the most unpleasant k i n d of move especially in time trouble.

to

face in such positions,

35 . . . QeS 36 Qd3 Be7 37 N£3 ( D iagram 3 2 ) 37 ... Nd7 3S Nh4 Nf8 39

N gf5 Qd7 4 0 Qg3 BdS 4 1 h6 NeS 42 N g 7 1-0

42 Nxg7 43 Nf5 is d evastating. Not a terribly accurate or th eoreti­ game (though I did promise Mark Ferguson to men­ . . .

cally s i g n ifi cant

tion that it won a best game p rize), but this isn't that kind of line. However, what the game does illustrate well is the sort of passivity Black must avoid, a n d h ow to go about it.

73

S t a rt i n g O u t : A l e k h i n e ' s D e fe n c e

Statistics Not good: White h a s scored 63% after 7 .. . N8d 7.

Theoretica l? N o t enormously; there i s noth ing which has to be learned i n t h e ex­ change-type lines .

Th eoretical Status White seems t o be a little better in t h e main variation, although a player who does not mind cramped positions and has a n eye to ma­ noeuvring for the counter-attack might like Black's game. Game 16 D Rogovski • Mu rashko Alushta 2001

1 e4 Nf6 2 e5 Nd5 3 d4 d 6 4 N£'3 Bg4 5 B e 2 c6 6 Ng5 ( D iagram 33)

Diagram

33

White accepts the challenge

Diagram 34 Black's turn to attack

An u nfashion able b u t quite dangerous move. White's point is th a t if 6 Bxe2 7 Qxe2 h e h a s h i s queen on its ideal square a n d has retained his k night rather than his bishop , which he hopes will b e to his a d­ vantage with th e fixed pawn structure that arises, a n d will also pre­ vent any accidents occ u r ri n g to the e-pawn. ...

6 ... Bf5 For the reason just given Black nor m a l l y prefers this move, which at re q u i re s White to hike his queen to d3 or g4 if he wants to force the exchange of light-squared bishops by 7 Bd3 or 7 Bg4 . Instead White normally prefers to sacrifice. least

74

Main L i n e w i t h 4...Bg4

7 e6

After 7 Bd3 Bxd3 8 Qxd3 Black's best is probably 8 . .. e6 rather than the older 8 . . . h6, when 9 Ne4 is dangerou s. 7 ... fxe 6 8 g 4

8 Bh5 + g 6 9 g4 Bxc2 10 Qxc2 gxh5 1 1 Nxe6 Qd7 is sim ilar but not quite so good for White. 8 . .. Bg6 9 Bd3 Bxd3 1 0 Qxd3 Nf6

Th e h- pa w n was attacked. 10 . . . g6 1 1 Qf3 Nf6 is m uch the same as the text. 1 1 N x e 6 Qd7 1 2 Q e 2 Kf7 1 3 Ng5+ Kg8 1 4 Rgl

White can draw with 14 Ne6, w h en Black hardly has better than 1 4 . . . Kf7 . White h ad various l ess effective alternatives in the last few moves, b u t Black's pieces will generally go to the same squ ares, save that . . . h 7- h 5 m ay be po s s i ble if not discouraged by Rg l . 1 4. . . Na6 1 5 Nc3 R e 8

1 5 ... N c 7 i s possible, b u t after 1 6 Bf4 there isn't m uch difference . 16 f4

White has two better moves. One is 16 N e6, when 1 6 . . . Kf7 1 7 Nxf8 Rhxf8 1 8 g 5 N d 5 1 9 Nxd5 cxd5 2 0 Qh5+ i s probably a little better for White. Th e other is 16 Bf4, wh en 1 6 . . . Nc7 1 7 0 - 0-0 Ncd5 1 8 Nxd5 Nxd5 19 Bg3 e 6 2 0 Rde 1 i s definitely a bit better for White. 16 B f4 e5 17 dxe5 h 6 1 8 Nge4 Nxe4 19 Nxe4 Qf7 20 Bg3 dxe5 has been pro­ p o s e d , although White sti ll looks a little better to me after 21 a 3 . 1 6. . .e 5 ! ? (Diagram 34)

A bold breakthrough. 1 7 d x e 5 d x e 5 18 f5

18 fxe5 was the critical move. Clearly Black h a s dangerous play after something like 1 8 . . . Bc5, b u t only future tests will tell whether it's enough; e5-e6 might be an im portant resou rce. 1 8... h 5 !

Black m u st play with all his pi eces, even though h e is obviously tak­ ing a risk opening the g-file. 19 h 3

More thematic w a s 1 9 gxh5, b u t one c a n see why White might not h ave liked 1 9 . . . Bc5 20 Rg2. 19 . . . hxg4 20 hxg4 Bc5 21 Be 3

Now Black is going to b e better. I t w a s more chal lenging to move the rook but White i s already struggling for equality. 2 1 ...Bxe3 22 Q x e 3 Nb4 23 Q f2

2 3 Qe2 was better. White m ust have underestim ated Black's next. 23 ... Qd4 24 Rei

White could not save the g-pawn in view of the fac t th at 2 4 Qxd4

75

S t a rt i n g O u t : A l e k h i n e' s D ef e n ce

exd4+ drops a piece or so.

24 Nxg4 25 Qg3 Qe3+ 2 6 Qxe3 Nxe3 27 f6 gxf6 28 Nge4+ Kf8 2 9 Nxf6 Nexc2+ ...

Simplest was 2 9 . . . Rh2 3 0 Nxe8 Nbxc2+ 3 1 Rxc2 Nxc2+ 3 2 Kd 1 Ne3+ 3 3 Kc 1 Kxe8 3 4 Rel Ng4 35 Re4 Rg2 a n d Black stays two pawns ahead and wins. 3 0 Ke2 Nd4+ 31 Ke 3 Rh3+ 32 Ke4 Rh4+ Fritz's cool 32 . . . Re7 33 Rcfl Nd3 may still win, but I imagine the players were in time trouble, and Black has a lready made the psycho·

logical error of driving the king up the board to become active i n the event of any s i mpl ificatio n . 3 3 R g 4 Rxg4+ 3 4 N x g 4 K e 7 3 5 a 3 N a 6 3 6 K xe5 R d 8 3 7 R h l Nf3+ 38 Kf4 Rf8+ 39 Ke3 Ng5 40 R h 5 'h-'h

Black could certainly have tortured his opponent a little longer, but he was no doubt happy to draw a gainst a higher-rated player.

Statistics White has scored 60% after 6 Ng5 .

Th eoretica l ? A l ittle. Black i s treading a narrow path with the king o n g8 and the rook on h8.

Th eoretica l Status Black's having a l ittle trouble a t t h e moment. H e needs t o do better rou nd about move 1 6, but below international level the line should remain quite playable . With 5 c6 as a whole Black h a s scored 42%. Each of the three lines shown here is presently givin g Black a headache, a n d the system is . . .

out o f fashion after a period in the sun during the late eighties a n d early nineti es.

76

C ha pter Th re e

M ain L ine w ith 4 . . . g 6/4 . . . Nc 6

R R R

B la c k P l ays 4 g 6 . . .

B l ack P l ays 4 Nc6 . . .

I l l ustrative G a m es

S t a rt i n g O ut : A l e k h i n e ' s D e f e n c e

Black Plays 4 g6 . . .

1 e4 Nf6 2 e 5 Nd5 3 d 4 d6 4 Nf3 g6 ( D i agram 1)

Diagram 1

Diagram 2

Black prepares to hit e5

Risky business

This is a very logical co ntinuation which came into the picture in the 1 960's a n d was su bsequently the particular baby of the great Rus­ sian/Am erican Alekhine devotee, Lev Albu rt. Black intends to assa u l t e5 with h is fianchettoed bishop. I f Wh ite succeeds i n m a in t a i n i n g this point the bishop is in danger of being shut out of the ga m e, so play can often become sharp as Black tries to break White's grip. In fact som e of the crucial variatio ns are more or less pure tactics. 5 Bc4

Almost u niversal now. 5 c4 �b6 will merely transpose to a . . . cxd6 Ex­ change variation, while the older 5 Ng5 c6 is ineffective, for exam ple 6 Qf3 f6 7 e6 Qa5 + 8 Bd2 Qb6, or 6 f4 Bg7 - White's f2- f4 plan i s less ef­ fective if he h asn't provoked . .. d 6 · d 5 or ... e7-e6 as h e does i n G am e 1 7 . 5 Ng5 Bg7?? 6 Qf3 is, however, to be a voided. 5 Nb6 •••

5 . . . c 6 is a lso playable a n d w a s dealt with i n Cha pter 1 , since it is b asi­ cally a possibl e introduction to the Kengis System. 6 Bb 3 Bg7

This is the modern starting point. White h as a choice of three methods: 7 Ng5 (Game 1 7), which aims at establishing his centre with f2-f4, 7 a 4 (Game 1 8) which aims, by forcing ... a 7-a5 , t o improve White's chances in the other vari ations, and 7 Qe2 (Ga me 1 9), which aims to deve l o p and maintain the e5-pawn with pieces rather than pawns. Of course si mply castling is a lso possible, but 7 Q.Q 0-0 8 Rel Bg4 9 h3 Bxf3 1 0 Q xf3 Nc6, for exam ple, i s already nice for Black - White's play looks logical, but he needs a more specific plan to maintain his centre.

78

M a i n L i n e w i t h 4 . . . g6/4

...

Nc6

Black Plays 4 Nc6 . . .

1 e 4 N f6 2 e 5 N d 5 3 d 4 d 6 4 N f3 N c 6 (Diagram 2 ) 4 . . . N c6 is a specialist's move. Any theoretical m a nual will tell you it's dubious. But there are players, notably Jurgens Fleck a n d Haakert from Germany, who h ave played it a lot and h ave achieved excellent results with i t - in fact if either h a s ever lost in t h e main variation (after 6 . . .fxe6) over the board then neither the d atabase nor their ow n extensive article reveals the fact. Th e reason is that this variation leads not just to sharp play but to very unusual positio ns. I n all probability, if Wh ite w a n ts an advan­ tage he has to go in for 5 c4 Nb6 6 e6 fx e6. White takes advantage of the knight on c6 (6 Bxe6?? 7 d5) to d amage Black's kingside struc­ ture a n d entomb the bishop on c8. This line stems from the earliest d ays of Alekhine's Defence. It h a s a lways been supposed to be good for Wh ite, b u t it remains today a minefield where even strong Wh ite players fear to tread. . . .

Before we get o n to that, h owever, let us consider what other ap­ proach White could take. Really t h e re are two. The first is to tra ns­ pos e into some kind of Exc h a n ge variation. For example, White could play 5 c4 Nb6 6 exd6 exd6 (6 . . . cxd6 7 d5 Ne5 8 Nd4 i s not such a good idea, but 6 . . . Bg4!? i s a n amusing gambit). White can then simply pro­ ceed as in Cha pter 5 with 7 Be2 Be7 8 0-0 0-0. Altern atively, he can go in for 7 d5 N e5 8 N d4. Perhaps Black should not let this go unchal­ lenged, but instead reply 8 . . . c5 9 dxc6 Nxc6, when his sl ightly better development will allow h im to get in . . . d6-d 5 and equ a l ise. Still on the same th eme, W h ite can be a bit more cun ning and try 5 c4 Nb 6 6 Nc3,

delaying the exchan ge for a moment. Black should simply proce ed with 6 ... Bg4. I f White now maintains the tension with 7 Be2 he is beginning to b e a t risk o f either 7 ... Bxf3 8 Bxf3 dxe5, or 7 ... dxe5 8 d5 Bx£3 9 Bxf3 Nd4, which forces unexplored gambit play on Wh ite, so probably he will u sually prefer 7 exd6 exd6, and now the game is a lmost certain to tra nspose to Cha pter 5 . 8 Be2 Be7 9 d5 Bxf3 10 Bxf3 Ne5 1 1 Be2 is a lso possible a n d i s supposed to give Wh ite a slight ad­ vantage, a lthough Black's plan of ... Bf6, ... a7 -a5, ... Ned7 -c5 , ... ReS a n d so o n gives him a very sol id position. It's not every day you see a 2500+ rated Russia n teenager fall for Le gal's Legacy, albeit with a modern twist, so I can't re s i st quickly showing the game Zhenkin­ Dyachkov, Orel l 996, which also il lustrates why 7 cxd6 m a y be doubtfu l . After t h at move t h is continued 8 B e 3 e6 9 h3 Bh5 10 d5 exd5 1 1 cxd5 Ne5 1 2 Bb5+ Nb d 7? ( 1 2 . . . N e d 7 w a s presu mably forced) 13 Nxe5!! Bxd 1 14 Bxd7+ Ke7 15 Nc6+! (th e twist) 1 5 . . . bxc6 16 dxc6 (threatening Nd5 m ate) 16 . . . Qa5 1 7 Rxd 1 Rd8 (Zhenkin does not comment o n th is; obviously Black has other ways o f staggering on, but h e is i n dire straits in a ny case) 18 b4 Qxb4 19 0-0 Qxc3 20 Bd4 Qc4 2 1 Rfe 1 + Qe6 22 Bxe6 fxe6 23 Rbl and White won. . . .

79

Start i n g O ut : A l e k h i n e ' s D ef e n c e

Alterna tively, White sometimes tries 5 Bb5. The simplest approach then is 5 . . . a6 6 Bxc6+ bxc6. 5 Bb5 is qu i t e frequently chosen by gra ndmasters who don't fa ncy a theoretical debate on the territory of lesser o pponents, but even so it's h ard to believe Wh ite c a n claim any advantage after giving up h is better bishop and strengthening Black's centre like this. Black usually develops with . . . Bg4, . . . e7 -e6, . . . Be7, . . . 0-0, . . . Nb6 (in reply to c2-c4) . If circum stances then allow him to force exd6 cxd6, u ncrumpling his pawns (p erhaps with . . . Nd7 if White h a s foolishly devel o ped his bishop o n e3), all well and good. Otherwise he can consider . . . a5-a4 (be careful, though , of . . . a 7 -a5 being met with a 2 -a4 a n d the a-pawn dro pping off later; Black must be in a position to m eet a 2- a4 with some concrete operation), or . . . d6-d5 followed by . . . c 7-c5 and/or . . .f7 - f6 to dismantle White's centre. There's not much theory on these variations and Black can develop a ll his p i ece s with­ out any immediate problems, so I h aven't given an illustrative game. 5 Bc4 Nb6 6 Bb5 a6 (6 ... Bd7 looks like a plausible k ind of move to me, though it hasn't been tried) leads to a similar game - White claim s to have gained by driving the knigh t from the centre first, while Black says he's gained a tempo because Wh ite will play c2-c4 anyway and the kn ight won't be en prise. Black should, h owever, give 5 ... d xe5 6 Nxe5 Nxe5 7 d xe5 Nb6?? 8 B xf7+ a miss, although 7 . . . c6 followed by . . . Be6 a n d . . . Nc7 is rea sonable.

The reason why White is u nder pres su re imm ediately after 4 . . . Nc6 is that a move l i k e 5 Be2 is met by 5 . . . dxe5 6 Nxe5 N xe5 7 dxe5 Bf5. Once a pair of m inor pieces has been exchanged and h e has ac tivated his queen's bishop this structure is comfortable for B lack (he should play . . . e 7 -e6 and not . . . g7 - g6) , especially if he can also exchange queens (White's e-pawn can easily prove difficu lt to defend). On the other hand 5 Be2 dxe5 6 d xe5 Bf5 is even less des irable since Black can al ready consider ... Ndb 4 with a n n oying t h reats against c2. I n sho rt, 4 ... Nc6 would b e what they w e re a l l pl aying if it w e re not for the repu tation of the pawn sacrifice proposed by Rubins tein i n 1 9 25 .

Illustrative Games Game 1 7

0 Gdanski • Ma rinkovic Biel 1 989

1 e4 Nf6 2 e5 N d 5 3 d4 d6 4 N fJ g 6 5 Bc4 Nb6 6 Bb3 Bg7 7 Ng5 (Diagram 3) 7 ... e6

The older 7 . . . d5 8 f4 0-0 9 0-0 f6 is u nder a cloud. White is always go­ ing to be a little better if he m aintains the pawn wedge on d 4 a n d e5 against the g7-bishop, and Black has no way to chal lenge it effec­ tively. The ga m e Karpov -K.Grigorian, Leningrad 1 97 1 w a s a particu­ larly impressive d e m on s tra tio n of White's play in this line. Wh ite can ins tead play 8 0-0 0 -0 9 Re 1 , intending to m eet .. . f7-f6 by exchanging and playing Ne6, w h ich also secures him an edge.

80

M a i n L i n e w i t h 4 . . . g 6/ 4 . . . N c 6

Diagram 3

Diagram 4

Here we go . . .

How would you exploit f6?

8 Qf3

This sharp move more often than not leads to a t e di o u s e n d ing, but Black has to know his onions against more violent tries. 8 f4 leads to pos itions very similar to those d iscussed in the next game. 8 ... Qe7 Neither 8 0-0 9 Q h 3 h6 10 Ne4 (see the classic positional crush Dzindzihashvili-Alburt, New York 1 984) nor 8 . Qd7 9 Ne4 dxe5 10 dxe5 Bxe5 1 1 Nf6+ is palatable for Black. ...

. .

9 Ne4

White h a s other moves, of course, but i f he doesn't go i n for t h e text then 8 Qf3 was just a stupid move, obstructing the f- pawn . 9

...

dxe5 10 B g 5 Qb4+

Exiling the queen l ike this looks ra s h b u t experience has shown tha t Bl ack m u st keep her in touch w i t h e5. ,

1 1 c 3 Qa5 (Diagram 4 ) 12 Nf6+

The tedious e n d in g I m e ntioned arises after 1 2 Bf6 Bxf6 13 Qxf6 ( 1 3 Nxf6+ Ke7 achieves nothing a n d loses a pawn) 1 3 . . . 0 -0 14 Qxe5 Qxe5 15 dxe5 Bd 7. Alburt c l a i m s this is equal, but in practice White has of­ ten gro u nd Black down from here with his space a dvantage (Khal if­ m a n-Fedorowicz, Wijk a a n Zee 1 9 9 1 and Ti m m a n-Alburt, Taxco 1 985 being two examples) . Instead Black might try 14 . . . Nc6. The subse­ quent ending after 1 5 Qxa5 is distinctly preferable to that a fter 14 . . . Qxe5. 15 Nf6+ Kh8 16 Qf4 Kg7 gi ves an in teresting position which is okay for Black. The problem is s u p posed to be 1 5 Qxc7 Nxd4 16 0 - 0 Nxb3 1 7 axb 3 Qxa l 18 Nf6+ as in a 1 985 Ljubojevic-Alburt game. This is meant to give White a winning attack after 18 . . Kh8 1 9 Qe5 Qxb2 2 0 g 4 Qc2 2 1 c 4 (threaten ing c4-c5) 2 1 . . . Qxb3 2 2 N d 2 Qa2 23 N de4, b u t presu mably doesn't since a good many gra ndmasters who were certainly not ignoran t of this piece of theory have elected to .

81

Start i n g O u t : A l e k h i n e ' s D e fe n ce

p l a y 14 ... Nc6 in the last twenty years, some of them no doubt after good chats with their computers. Unfortunately they haven't been kind enough to publish their idea, so we'll have to guess, and I'm afraid after a bit of trying I don't have a guess. (Black can decline the rook with 1 7 ... Qf5, of course, b u t he must be a bit worse then.) 1 2 Kf8 ...

Evidently the only move. The position now is viciously tactical. 13 d5

The only sensible way to meet Black's threat of . . . exd4, for example 1 3 0 - 0? exd4 1 4 Qf4 N8d7 1 5 cxd4 Qf5. The text sidelines Black's queen. 13

...

e4! (Diagram 5)

D i a g ra m 5

Diagram 6

Black stays alert

The smoke has cleared

Again this is pretty much forced as 1 3 . . . exd5 1 4 Nxd5 Nxd5 1 5 Bxd5, for example, is crushing. The point of t h e text is that 14 Qxe4? drops a piece a fter 14 . . . h6 1 5 dxe6 Bxe6, as Joe Gallagher found out the hard w ay in Gallagher- Blees 1 988, although with typical resilience he didn't let that stop him drawing the ga me. 1 4 Nxe4

Understandably avoiding the complications which a rise after either 14 Qg3 or 1 4 Qf4, but the text enables Black to consolidate. Mter 1 4 Qf4 Black ca n obtain good play with 1 4 . . . Nxd5! 1 5 Bxd5 h6! ( 1 5 ... exd5? 16 b4 is dire) 1 6 Q x e 4 (the point of Black's clever inter­ mezzo was that 16 Bh4 exd5 1 7 b4 g 5 is now possible; 16 Bxe4 hxg5 1 7 Qf3 Qe5 18 Ng4 Qf4 is a l so good for Black) 1 6 hxg5 ! 17 Bxb7 Rh4 , w h e n either 1 8 Q d 3 Bxf6 1 9 Bxa8 Ba6 20 Qe3 N d 7 or 1 8 g4 Bxb7 1 9 Qxb7 Bxf6 2 0 Qxa8 Qe5+ and 2 l . . .c6 gives Black i nteresting play for his exchange. . . .

Nigel Short's 14 Qg3 is critical, when 1 4 . . . Na6 is the best move, and after 15 dxe6 either 15 ... h6!? 16 e7+ Kxe7 1 7 Nd5++ Ke8 18 Nxc7+ Nxc7 1 9 Qxc7 Qxg5 2 0 Qxf7+ Kd8 21 Qxg7 ReS, when Black has sur-

82

M a i n L i n e w i t h 4 . . . g 6/4 . . . N c 6

prisingly good com pensation, o r 1 5 . . . Qf5 1 6 e 7 + Kxe7 1 7 Ng8++ Ke8 1 8 Nh6 Bxh6 19 Bxh6 Be6 20 0-0 Bxb3 2 1 a xb3 f6, when White h as unsurprisingly good compensation, are unresolved. 1 4 ... exd5 15 0-0

I n a later game another opponent of Marinkovic tried 1 5 Be3 N8d 7 1 6 Bxb6? axb6 1 7 Ng5 N e 5 1 8 Qg3 h 6 and lost considerably more quickly. 1 5 . . . Be6

And not, of course, 1 5 . . . dxe4?? 1 6 Qxf7 m ate. 16 Ng3

An admission of defeat. Black now has a n extra pawn and merely needs to develop and consolidate, but 1 6 Nf6 N8d7 j us t promotes ex­ changes. 16 ... N8d7 17 Nd2 ReS 18 Bc2 Ne5 19 Q f4 N b d 7 20 Bh6 Qb6 2 1 Rael B x h 6+ 2 2 Qxh6+ Kg8 2 3 h 3 f5 2 4 Nb3 Nf7 (Diagram 6)

Expelling the queen. Black has consolidated and is now wi n n in g al­ though the process takes a long time. ,

2 5 Q d 2 Nf6 26 Nd4 Bd7 27 Rxe8+ Bxe8 28 R e l B d 7 29 h4 Kg7 3 0 h5 R e 8

White a lways s e e m s to b e a move behind - h e g o t Nd4 i n j u s t when Bl ack had a rran ge d to retreat the e6-b ishop, and now he is one move too late to confine the black king and rook i n the corner with h 6 +. 3 1 Rxe8+ Nxe8 32 h6+

Th is m erely creates a weakness. I suspect White would have done better to try 32 hxg6, a lthough he must be lost. 32 ... Kf8 33 Bb3 Qd6 34 Nf3 Nf6 35 Ne2 c6 3 6 g3 Ng4 37 Kg2 Qc5 3 8 N f4 b 5 3 9 Qe2 Qe7 4 0 Qd2 N fxh6

I don't see why 40 . . . Ngxh6 41 Qd4 Bc8, preserving the a -pawn, wasn't possib le, but m aybe Bl ack was in time trouble. 41 Qd4 Nf7 42 Qxa7 Be8 43 Qd4 Qe4 44 a4 Qxd4 45 Nxd4 bxa4 46 Bxa4 Nd8 47 Nd3 Ke7

Black h as m ad e thi n g s di fficult for h i mself, but h e kn uckles down a n d gri nds i t o u t. 48 f3 Nf6 49 Krl Kd6 50 b4 B d 7 51 Nc5 Bc8 5 2 Bc2 Ne6 53 Ncxe6 Bxe6 54 B b l Bc8 55 Bd3 Bd7 56 Bc2 Nh5 57 f4 Nf6 58 Nb3 Bc8 59 Bd3 Ne4+ 60 Bxe4 dxe4 6 1 c4 Be6 62 Nd2 c5 63 b5 Kc7 Black is winning now. There is no defence to the plan of . g5x f4 a n d then using the h-pawn to divert White's king. .

.

h 7 -h6,

. g6. .

64 Ke3 Kb 6 65 Ke2 h6 66 Ke3 Ka5 67 Krl Kb4 68 Ke3 Bc8 69 Kf2 B b 7 70 Ke3 Kc3 7 1 g4 h 5 7 2 gxh5 gxh 5

White h a s done Black's work of creating a passed pawn for him, but it made no difference. 73 Ke2 h4 74 b 6 h 3 75 Nfl Kxc4 7 6 Ne3+ K b 5 77 Nxf5 h2 78 Ng3

83

S t a r t i n g O u t : A l e k h i n e ' s D e fe n ce

Kxb 6 7 9 f5 K c 7 8 0 Kf2 e 3+ 0-1

Statistics White has scored 60% a fter 7 Ng5 but only 54% a fter the subsequ e n t 7 . . .e 6 .

Theoretica l? Brutally s o , I fea r. The position a t move 1 2 i n vol ves virtu ally nothing a n d i f you don't k now them you'll have to work them out ­ good luck !

but tactics,

Theoretical Status Not clear. There hasn't been a l o t of practice in t h i s line in the l a s t t e n years, a n d w h a t there h as b e e n h as m ainly b e e n in t h e endi ng. I f Black's h a ppy t o grind t h a t o u t t h e n theoretica lly h e is fine. I f h e wants to sharpen t h e p l a y then everyth ing turns o n 1 4 ... N c 6 - i f Black ca n avoid getti ng mated in the style of Ljubojevic-Albu rt then he can look forward to reasonable play there, and the sharp varia­ tions i n the notes to the game give Black fu ll-blooded cha nces. Game 18 D Grischuk • P onoma riov Torshavn 2000

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 g6 5 Bc4 Nb6 6 B b 3 Bg7 7 a4

White a i m s to provoke . . . a 7-a5. This h as various benefits for him - if he i ntends to play the line in the game then Black's knight on b6 i s unprotected, wh ich makes life that little bit harder for Black i n the sharp positions which a rise, while some other points are explained in the next game. 7 ... a5

A very critical moment. What i f Black doesn't cooperate? The obvious non-cooperative move is 7 . . . dxe5 , and now 8 Nxe5 Bxe5! 9 dxe5 Qxd1 + 10 Kx d 1 Nc6 1 1 a 5 Nd7 1 2 e6 fxe6 1 3 Bxe6 Nde5 is comfortable for Bl ack. White, however, h as a cunning p l a n , which i s 8 a 5 N6d7 9 Bxf7+ (already there is n o way back: 9 Qe2 0 - 0 1 0 dxe5 Nc5 and 9 dxe5 Nxe5 10 Qxd8+ Kxd8 1 1 Ng5 Ke8 a re both fine for Black) 9 . . . Kxf7 1 0 Ng5+ Kg8 1 1 N e6 Q e 8 12 Nxc7 Qd8 (this is the best square, preventing the knight getting out from a 8, although 12 . . . Qf7!? has also been played) 13 Nxa8 ( 1 3 Nc7 draws by repetition, which was goo d enough for Peter Svidler i n Svidler-Khalifman 1 995, u n less Black fa ncies Qf7!? next time around) 13 ... exd4 ( D iagram 7 ) . . .

This position i s critical for 4 . g 6 a n d i s very complicated a n d in teres t­ ing to analyse. White's first try was 14 c3 Nc5 ( 1 4 . . . Nf6 15 cxd4/0- 0 Na6 i s another notion, just allowing White to recapture on d4, but us­ ing the extra time to develop and win the aS-knight) 15 cxd4 ( 1 5 b 4 .

84

.

M a i n L i n e w i t h 4 . . .g6/4. . .N c 6

N e4 ) 1 5 . . . Bxd4 1 6 0 - 0 , but after 1 6 . . . e5 17 Be3 N b a 6 (further corral­ lin g the knight) 18 Nc3 Bf5 1 9 Rc1 Kg 7 20 Bxd4 Qxd4 Alburt picked up the knight a n d won in Ch risti ansen-Albu rt, USA 1 990. In this line 16 Be3 has also been tried and got nowhere: 16 ... e5 17 Bxd4 exd4 en­ ables White to force the e - p a w n to d4, where he can at l east w i n tha t in retu rn for h i s aS-knight, b u t n o more, for example 1 8 0 - 0 N b a 6 1 9 Nd2 Be6 2 0 Re 1 Bf7 2 1 N e4 d3 2 2 b4 Nxe4 23 Rxe4 Kg7 2 4 Ra3 d2 2 5 Re2 Re8 2 6 Rxe8 Bxe8 2 7 Ra2 Q x a 8 28 Rxd2 (Oral-de Firmian Reyk­ javik 2 00 0).

Diagram 7

Diagram 8

A key position

Tricky

Recently, therefore, returning to dia gram 7, White has tried 1 4 0-0 Nc6 ( 14 ... Nf6 1 5 c3 Na6 is the s a m e a s a fter 14 c3 Nf6) 1 5 c3 Nc5 1 6 b4 dxc3 1 7 Qxd8+ N xd8 1 8 bxc5 ( 1 8 Ra3 has been s u gge s te d , when I think Black ought to play 1 8 . . . c2) 1 8. . . c2 1 9 Nc3 (the only move to prevent greater m a terial losses) 1 9 . .. Bxc3 20 Ra3 (or 20 Ra2 Bf5 2 1 Nc7 when I suggest 2 1 . . .Nc6 i n place of 2 l . . . Kf7, Blehm-Nava rra, Av­ iles 2000) 20 . . . Bf6 2 1 Nc7 Kf7 22 Nd5 Nc6 23 Nxf6 exf6 24 f3 h 5 , when Black had reasonable compensation in Almasi-de l a Villa Gar­ cia, Pam plona 1 99 9 . A t t h e m oment these gam es, which h a v e been annotated in various places, represent the last word in theory, but clearly there are many other moves both sides could try . If Black can't make this work then his whole opening scheme is in trouble, since the present game is dis­

couraging indeed. 8 Ng5

Apart fro m t h i s 8 exd6 is also q u ite annoying. Bl ack m i ght be reduced to 8 . . . Qxd6, since 8 . . . exd6 does not go with . . . g7 -g6 and 9 0-0 0-0 10 Bg5 is good for White, while 8 . . . cxd6 9 0-0 0-0 10 h3 is also nice for White, whose bishop is safe fro m . . . N a 5 a nd who h as the splendid b5outpost. 8 . . . e6 9 f4

85

S t a rt i n g O u t : A l e k h i ne ' s D e fe n c e

9 Qf3 is no t s o po p ul a r with 7 a4 a5 in , sin ce now 9 . . . Q e 7 1 0 Ne4 dxe5 1 1 Bg5 Qb4+ 12 c3? drops the bishop o n b3. Instead White h a s to try 1 2 Nbd2 when Black should not play 12 . . . Qxd4? 13 0 -0-0 with a strong in itiative, but 1 2 . . .f5 1 3 Nf6+ Kf7 with entertaining and virtu­ ally un tried com plications. 9 . . . dxe5 10 fxe 5 c5 1 1 c 3

Less good i s 1 1 0 -0 Qxd4+ 1 2 Qxd4 cxd4 1 3 Nxf7 0 - 0 . 1 1 . . .cxd4

Black needs to make this exchange s i n ce after 1 1 . ..0 -0 1 2 0-0 Nc6 White can ignore the threat to the d - p awn with 1 3 Ne4! and slaughter Black as in the famous game Kas parov-Palatnik, Daugavpils 1 9 78 one of the games whose exubera n t tactical genius first a n nounced to the chess world that 1 5-year-old Garry Kas p arov was a c ha m p i on in waiting. 1 2 0-0 (D iagram 8)

White uses the tactic 12 . . . dxc3?? 13 Qxd8+ a n d Nxf7+ to force 1 2 . . . 0 -0, although I don't see that 12 cxd4 Nc6 13 Nf3 f6 1 4 Nc3 would actually h ave any concrete d rawback. 12 . . . 0-0 1 3 cxd4 Nc6 14 Nf3 f6 1 5 Nc3!

A very strong p a w n sacrifice. 1 5 exf6 Qxf6 1 6 Be3 N d 5 1 7 Bf2 Nf4 1 8 N c 3 Nh3+ i s the old line. I f White wanted to play this he shouldn't h ave included 7 a4 , since now after 19 gxh3 Qxf3 20 Qxf3 Rxf3 h e ha s to spend time defending t h e bishop on b3, b u t in either event Black does a lot better there than he does a fter the text. 15 ... fxe 5

Black could play 1 5 . . . N d 5 b u t i f h e does h e i s just conceding t h a t he's going to be worse a t no material cost. 16 Bg5 (Diagram 9)

86

Diagram 9

Dia gram 1 0

The point

A fitting finish

M a i n L i n e w i t h 4 . . . g 6/4 . . . N c 6

1 6 Qd7 •..

16 . . . Qe8 17 dxe5 Nxe5 18 Nxe5 Rxfl + 1 9 Qxf1 Bxe5 20 Re 1 is pretty much the same, when White h a s trem endous com pensa tion with the threat of N e4 -f6. 1 7 d x e 5 Nxe5 18 Nxe5 Rxfl+ After 18 . . Qxd 1 19 R a x d 1 Bxe5 White still has an extremely strong initiative for the pawn and Black is struggling to develop his queen· side pieces, for example 20 Rfe l Bg7 21 Be 7 Rf5 22 g4 Rf4 23 Bc5 was already winning i n one gam e. .

19 Qxfl Qd4+

19 . . . Bxe5 20 Rd 1 , with the i d e a o f RdS+, wins at once for Wh ite. 20 Khl Qxe5 2 1 Bd8! Drawing Black's a ttention to the merits of interpol ating 7 a4 a 5 in the most brutal way. Without these moves the position would be n o better than u nclear for Whi te.

2 1 . . . Qc5

Th is h a d all ha ppened before in a couple of games, and Black was supposed to be okay after 2 l . ..Ra6. In fact he is a nything but okay as Volzhin demonstrated that 22 Rd 1 is winning for White. He is threat· ening 23 Bxb6 Rxb6 24 Rd8+ and nothing m uch helps Black, for ex· am ple 22 . . . Na8 23 N d 5 !, 22 Qf5 23 Qxf5 gxf5 24 Bxb6 Rxb6 25 Rd8+ and Nb5 winning a piece, or 22 . . . h 6 23 Bc2. Volzhin suggests that if Black wants to keep playing 2 1 . . . Qd6 22 Rdl Nd5 is the only way, b u t his 2 3 B h 4 ! is still very strong. Fritz's 23 . . . Bxc3 24 bxc3 B d 7 i s not a s simple to refu te a s i t looks - Black can a t least develop a l l his piece s ­ but I a m sure White is much better, while 2 3 . . . Qf8 24 N x d 5 Qxfl + 2 5 R x f1 exd5 2 6 Bxd5+ K h 8 2 7 b3 i s also horrid. . . .

22 Ne4 Qb4 2 3 Ng5 Kh8 2 4 Qf7 Bd7 2 5 B x e 6 R x d 8

2 5 ... Bxe6 2 6 Nxe6 also m ates. 26 Qg8+ Rxg8 27 Nf7 mate (Diagram 1 0)

Black didn't look like someone who would be world cham pion two years later in this game, but presumably he relied upon theory until it c a m e tim e to play 2 l . . .Ra6, noticed 22 R d 1 , and after that things are grim .

Statistics Black h a s managed 2 0 % after 1 5 N c 3 , though he does okay i f White refrains from this critical line.

Th eo retica I? Fairly - o n e could

p u n t 7 . .. dxe5 a n d t h e excha nge sacrifice and hope work it out, but the m a i n thing to rem ember is to a void the position after 8 Ng5.

to

87

S t a rt i n g O ut : A l e k h i n e ' s D ef e n c e

Theoretica l Status Th e l ast word h a s yet to be said o n t h e exchange sacrifice. I f that doesn't work Black needs a fundamental new idea quite early i n this l ine to keep 4 . . . g6 alive a s an independent system. Game 1 9 D V.Dimit rov • Ma rinkovic

Tivat 1 995

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 g6 5 Bc4 Nb6 6 Bb3 Bg7 7 Q e 2

White's other m oves a r e not particula rly effective. 7 N b d 2 was Spassky's choice, but Fischer's 7 . . . 0-0 8 h3 a5 9 a4 dxe5 10 dxe5 Na6 (remember this manoeuvre) is already nice for Black. 7 e6 Bxe6 8 Bxe6 fxe6 9 N g5 Qd7 1 0 Qf3 Bf6 is poor. 7 0 - 0 Nc6 8 Qe2 is the note to White's 8th. 7 ... N c 6

This is not Black's only id e a . He ca n also p l a y 7 ... 0 - 0 8 h 3 (again White should prevent 8 ... Bg4 ) 8 . . . a 5 9 a4 dxe5 10 dxe5 Na6 1 1 0-0 ( 1 1 Be3 Nd5 1 2 0 - 0 Nxe3 1 3 Qxe3 gives White nothing m uch) 1 1 .. . Nc5 1 2 R d 1 Qe8 1 3 Nc3 Nxb3 1 4 cxb3. I t looks a s though this pos ition ought to be quite nice for Black , but in fact it i s very difficult to find much to do. 1 4 . . . Be6 1 5 N d4 Bd5 16 Nxd5 Nxd5 17 Nf5 (Nunn- Howell, Port Erin 1 994) doesn't necessarily have to be bad for Black, but neither 17 ... c6 18 Nxg7 Kxg7 nor 17 ... gx f5 1 8 Rxd5 is exactly comfortable for him . 8 h3! White a voids t h e l i n e 8 0 - 0 d x e 5 9 d xe5 N d4 1 0 Nxd4 Qxd4 , which i s fine fo r Black since 1 1 Re 1 Bg4 i s disruptive, and 1 1 e 6 B x e 6 1 2 Bxe6 fxe6 m isses the ma rk after either 13 Qxe6 Qc4 o r 13 N d2 (to control c4) 13 . . . e 5 . If Black now tries 8 ... dxe5 9 dxe5 N d4 1 0 Nxd4 Qxd4 th en the point is that 1 1 f 4 is legal, followed by Nc3, Be3 and so on. 8 .. . a5 Black can also try 8 . . . 0 - 0 9 0-0 N a 5 when h e obtains the t w o bishops, although White h a s decidedly more s pace in a messy position a n d has

usually managed to retain an edge. After 10 Nc3 N xb 3 1 1 a xb3 Speelman's 1 1 .. . h6, preventing Bg5 a n d preparing ... f7.f6 and . . .fxe5, may b e best. If White is worried about this line t h en he can t ry to prevent it by 7 a 4 a 5 , sinc e ... Na5 is now i mpossible. 9 a4 0-0 1 0 0 - 0 dxe5

Of course this simplifying idea is not forced, but 10 ... Bf5 a n d espe­ cially 1 0 . . . d 5 have never done very well.

1 1 dxe5 Nd4 1 2 N xd4 Qxd4 13 Rei e6

Black can also try 13 ... Bd7 when 14 Nd2 e 6 is the game and 1 4 Nc3 e6 15 Bg5 h 6 equalises sinc e 16 Be7? Rfe8 1 7 Rad 1 Qxe5 wins Black a pawn. 1 4 c3 Qh4 is also reckoned to equalise, but I would have

88

M a i n L i n e w i t h 4 . . . g 6/4 . . . N c 6

thought that 1 5 e 6 Bxe6 1 6 Bxe6+ fxe 6 1 7 Qxe6 will always leave White with a nagging edge. 1 3 ... Be6, braving the doubled pawns, or 1 3 . . . Bf5 fol lowed b y . . . Qd7 and . . . Be6, are also possible. 1 4 Nd2 (Diagram 1 1)

Diagram 1 1

Dia gram 1 2

T h e second knight heads for f3

Black holds firm

14 . . . B d 7

In the earlier game Short-Timman, Tilburg 1 99 1 , Black had played 1 4 . . . Nd5 a t this point, which White had met with 1 5 N f3 Qc 5 16 Qe4 Q b 4 1 7 Bc4! (a brilliant idea, m utilating h i s queenside pawns and giv­ ing up the bishop just to get the queen to h4; meanwhile accepting Black's pawn sacrifice with 17 Bxd5 exd5 18 Qxd5 is b e neat h t h e con ­ tempt of a n y commentator I've seen - Black's bishops a nd light­ square control evidently give him compensation, but whether they give him ful l equ ality is perha ps another matter) 1 7 . . . Nb6 18 b3 Nxc4 1 9 bxc4. Stop m e if you've see n it before, but I can't resist showing the rest: 19 . . . Re8 20 Rd 1 Qc5 21 Qh4 b6 22 Be3 Qc6? (22 . . . Qf8 was esse n ­ t i a l according to Rogers; whether White i s actually then much better in this much -quoted game is open to debate) 23 Bh6 Bh8 24 Rd8 B b 7 2 5 R a d 1 Bg7 (26 Q e 7 was threatened) 2 6 R 8 d 7 R f8 ( t h e move Timm a n had overlooked u n ti l too l a te w a s 26 . . . Qe4 2 7 Rxf7!) 2 7 Bxg7 Kxg7 2 8 R 1 d4 (preventing Qe4) 28 . . .R a e 8 2 9 Qf6+ Kg8 3 0 h 4 h 5 , a n d now the entertaining 31 Kh2 ReS 32 Kg3 Rce8 33 Kf4 Bc8 3 4 Kg5 1 - 0 in view of Kh6 and ma te. . . .

15 N fJ Qc5 16 Be3 Qe7 17 Bg5 Qc5 18 Qe4

White coul d draw, of course, b u t how he is to m a k e progress is not clear. The text tries to get to h4 as in Sh ort-Tim ma n, b u t doesn't

really work. Meanwhile Black was ready to start uncurling with moves like . . . h 7- h 6, . . . Bc6 a n d . .. Rad8. 18 . . . Bc6 1 9 Qh4 Rfe8 ( Diagram 1 2 )

White w a s threatening 20 Be7.

89

S t a r t i n g O u t : A l e k h i n e ' s D e fe n c e

20 N h 2

Nothing w a s much better than this. M y computer loves 2 0 B e 3 Q e 7 2 1 Qxe 7 Rxe7 2 2 Bxb6 both i n this position and i n many i n this varia­ tion, b u t this leads nowhere as Black can regroup by doubling rooks on the d-file, taking his bishop to c5 to remove th e knight on d4 and playing his kin g to the centre. He should be careful about ... Bd5; ex­ changing these bishops may just give White's knight the eternal b 5 outpost, a n d White's k i n g i s already eyeing the c4-b5 pathway long­ ingly. 20 ... Qb4 2 1 Qxb4 21 Ng4 h 5 2 2 N f6 + Bxf6 2 3 Bxf6 Qxh4 2 4 Bxh4 Bd5 25 Bxd5 exd5 (and not 25 ... Nxd5 when White's domin a tio n of the d-fi1e will give him

the advantage) is also about equal - Black can continu e with . .. c 7-c5 , . . . d5-d4 a n d . .. N d 5 when his play on th e queenside is just a s good a s a nyth ing White c a n generate on t h e kingside. 2 1 ...axb4 2 2 a5 N d 7 23 Ng4 h 5 24 Nf6+ Nxf6 2 5 exf6 Bf8 26 Bf4 Bd6 2 7 Bxd6 cxd6 2 8 Red l d 5 2 9 Rd4 Ra7 30 Rxb4 %-% After Black regains his pawn with 30 . . .Rea8, there is little left to play

for.

Statistics White has scored 66% after 7 Q e2 .

Theoretica l ? Not desperately so. White plays for a quiet edge.

Th eoret i cal Status Black is a bit passive in t h e sys tem in the game, al though solid enou gh . I would be inclined to investigate the ... N a6 idea a bit more.

4 ... g6 in general needs a lot of work if Black is going to be able to play it with confidence . In my opinion Black si mply can't allow the system in Game 1 7 a n d he h a s to make Alb urt's exchange sacrifice work. As theory stands Black also has to accept fairly d u ll, albeit solid, posi­ tions in Games 16 a n d 1 8 . Game 20 0 Zuidema • Haakert Den Haag 1 985

1 e4 N f6 2 e 5 Nd5 3 d4 d6 4 Nf3 Nc6 5 c 4 Nb6 6 e 6 fxe 6 (D iagram 1 3)

6 . . . Bxe6? 7 d 5 wins a piece, of course. It's only fair to say that White has six moves in the dia gra m position, all of which are dangerous (7 d5, 7 Nc3, 7 Ng5, 7 Bd3, 7 h4 and 7 Be3, since you ask) and ma ny of which can transpose, confusingly, to each

90

M a i n L i n e w i t h 4 . . . g6/4 . . . Nc6

other, a n d t h a t a book l ike t h i s can't poss ibly give more than a very general view, especially since this is not a m ain line. Fortunately, though, Herren Fleck a n d Haakert were kind enough to share their secrets in a n extensive th eoretical article i n 'Ka issiber 19' - a w hole 20 pages packed with games a n d original a n alysis. (Kaissiber is a fine German fanzine specialising i n ch ess h istory and rare yet playable opening variations). If a nyone is seriously i n t e r e ste d in this line, then I recommend sending 7 . 1 0 Euros to the editor Stefa n Bucker without further a do. Having said th at, th is is a variation where experience and a ge ne r a l idea of Black's aims do he l p so it is possible to play i t a t a lower level without d etailed th eoretical knowledge. ,

Diagram 1 3

Diagram 1 4

Let the games commence . . .

Who i s better?

The variation which best illustrates Black's ideas is 7 N g5 g6 (7 . . . e5 is also poss ible) 8 Bd3 (8 Qf3 Nxd4 9 Qf 7+ Kd 7 10 Be3 Nc2+ 1 1 Kd2 Nxa 1 o r 1 l .. .Nxe3 is fi ne for Black, a s is E CO's 8 d 5 exd5 9 cxd 5 Nb4 10 Qf3 N4xd5, a lthough i n both cases great c om p l ica t io n s rem a i n and i n the l atte r case Wh ite ca n draw a t once with 1 1 Qf 7+ Kd 7 12 Qe6+ Ke8) 8 . . . Nxd4 9 Nxh7 Bg7 10 Bxg6+ Kd7 (Dia gram 14). This position was given a s clearly better for White without a n a lysi s by Hort a n d Sch warz, yet p r a c tice favours Black. Black ca n continue with ... c 7-c5 , ... Kc7, . . . Bd7, ... Bc6 and ... Qg8; all his pieces have pros­ pects, a n d Wh ite has weak pawns to defend. I n the l ong term Wh ite's king may be the less safe. Obviously the kingside h as its dangers, and on the queenside he can never e nj oy the cosy carapace of pa w ns th a t Black's monarch has. In the short term Black c a n meet 1 1 N g5 either by giving u p the exchange with 1 1 . . . Nxc4 or with 1 L . Q f8 - a fter the latter m ove V.Gurevich-llczuk, Koszalin 1 999 continued 12 Nc3 Nxc4 13 Qa4+ b5 1 4 Nxb5 Nb6 15 Qa5, when White was able to save h i m ­ self o n l y b y pulli n g o u t h is Grandm aster c a r d a n d offering a draw. 7 Bd3 can also be met with 7 . . . g6: 8 h 4 Bg7 9 h5 Nxd4 10 Nxd4 (or 1 0 hxg6 Nxf3+) 1 0 . . . Bxd4 1 1 h xg6 hxg6 1 2 Rxh8+ Bxh8 1 3 Bxg6+ Kd7 i s

91

S t a rt i n g O u t : A l e k h i n e ' s D e fe n ce

similar to the above and is not dangerous ; 9 Be3 e5 10 d 5 Nb4 11 Be4 (White m ust preserve this piece) 1 l . . .e6 i s a lso fine for Bl ack. 7 d 5 exd5 8 cxd5 Nb4 9 Nc3 e6 10 dxe6 Bxe6 leaves White struggling to demonstrate com pensation. 7 h4 is directed ra ther obviously a gainst Black's . . . g7 -g6. The traditional recipe is 7 . . . e5 8 d 5 N d4 9 Nxd4 exd4 1 0 Qxd4 e5 1 1 Qdl , when Fleck suggests 1 l . . . Bf5 1 2 Nc3 h 5 13 Bd3 g6 with the intention of . . . Bh6. This is possible, of cou rse, but I m ust say White looks a little better to m e. Anoth er p os s i bility is 7 .. . Nd 7, with the inten tion of taking this knight to f6 and continuing with . e6-e5, thus hopefully making h4 look stu pid . Finally, the defian t 7 g6 m ight be worth trying, for example 8 h5 Bg7 9 h xg6 h xg6 1 0 Rxh8+ Bxh8. Now White c a n collect the g-pawn with eith er 1 1 Qd3 o r 1 1 Bd3, b u t either o f those moves can b e answered nonchalantly with 1 l . . .Kd7, when play is si m il a r to variations given above, although White does retain h is d-pawn. . .

. . .

7 Be3 c a n lead to the position considered in the note to White's ninth in the game. Otherwise 7 . .. g6 8 h 4 Bg7 9 h5 (9 Nc3 is the transposi­ tion I m ention ed) 9 ... e 5 1 0 d5 e 4 ! 1 1 Ng5 N e 5 leads to splendid com­ plica tion s after either 12 h x g6 or 12 Nxe4 - these are analysed in de· tail in Kaissiber with the conclusion t h a t Black is okay. 7 Nc3 g6 8 h4 Bg7 (Diagram 1 5)

Diagram 1 5

Diagram 1 6

Should White push the h-pawn?

Safe castling?

9 h5

After considering the above these moves should need no explan ation, but this was an important mom e nt. Black is now ready for . . . e6-e5, d5 Nd4 and the best move is probably 9 Be3. Fleck says so, a n d the then soon -to-b e F I D E World Cham pion Ruslan Ponomariov thought so too when Iva nchuk played t h is va riation against him in their title ma tch. The idea is simply to meet ... e6-e5 with d5 Nd4, Nxd4 exd4, Bxd4. Nonetheless Black should continu e 9 . . . e5 10 d 5 Nd4 1 1 N x d4 exd4 1 2 Bxd4 Bxd4 1 3 Qxd4 e 5 1 4 dxe6 (or 1 4 Qe3 0-0 1 5 h 5 g5 1 6 h 6 Rf4 with

92

M a i n L i n e w i t h 4 . . . g6/4 . . . Nc6

sharp play) 14 ... 0-0 15 0-0-0 (15 h 5 Qf6 ha s seen several draws) 15 . . . Bxe6 16 Ne4 Q e 7 17 h5. Fleck gives White the advantage here, although I'm not sure this is so significant after 1 7 . . . Qg7. But there is no denying that this variation spoil s Black's fun ; probably White can retain a t l east a little edge this way. Another order is 9 ... 0-0 10 h5 e5, but the trouble then i s that White can flick in 1 1 hxg6, when 1 1 . . .exd4 12 gxh7+ is probably just a bit too dangerous, and i f 1 1 . . .hxg6 12 d5 Nd4 1 3 Nxd4 exd4 14 Bxd4 Bxd4 15 Qxd4 e5 16 Qe3 White has suc­ ceeded in opening the h-file al ready and stands a little better than a f­ ter 14 Qe3 above, although it's not necessarily a nything fatal; Black continues . . . Qf6 and . . . Qf4. 9 ... e5 10 d5 Nd4 1 1 hxg6

White can al so play 1 1 Nxd4 exd4 12 Nb5 e5 13 hxg6 Bf5 ! transposing to the game, save that 14 dxe6 is im possible. 1 1 . .. Bf5

1 1 . . .B g 4 (the most popular) a n d si m ply l l . . . hxg6 are also possible, with the usual complications. 12 N xd4 exd4 13 Nb5

Both 13 gxh 7 Qd7 and 13 Rxh7 Rxh7 14 gxh7 Kd7! produce perplex­ ing complications, but Black seems to be fine. 13 ... e 5 1 4 dxe6

1 4 Rxh7 Rxh7 15 gxh7 Kd7 is a t least okay for Black. 14 ... Qe7 15 Nxd4 Bxg6 16 Q g4

Ha akert had previously had a postal game which had gone 1 6 Be3 0 0-0 1 7 Qg4 R h f8 with a very sharp position i n which he felt Black stood well, although in that game he h a d outcombined himself and lost. The text prevents long castling on account o f 1 7 Bg5. 1 6 . . . 0-0 (Di agram 1 6) Cheeky, but White cannot get at the king now that he cannot trade bishops with Bd3, a n d in fact Black's rooks come to the ideal 'e' a nd f. files sooner this way. 1 7 Bg5 Bf6 18 Bxf6

H a akert had intended to meet 18 Bh6 with 18 . . . c5!? 19 Bxf8 Rxf8 20 Nb5 Bxb2 with great complication s, for example 2 1 Rd 1 (2 1 Rbl? Qf6 is strong) 2 1 . . . a 6 22 Nxd6 Bc3 + 23 Ke2 . It is understandable that h i s opponent w a s not u p for this. Quite possibly t h e position is objectively good for Black, a n d certainly one slip could spell the end for White. Instead White sends his king to safety and n a vigat e s the game to­ wards a draw. 1 8 ... Qxf6 1 9 0-0-0 Qxf2 20 Bd3 Qf4+ 2 1 Qxf4 Rxf4 2 2 Bxg6 hxg6 23 b 3 a6 24 Rhfl Raf8 25 Rxf4 R xf4 26 Rei Nc8 27 e 7

I f White were t o allow this pawn to b e blockaded he might easily stand worse. 27 ... Nxe7 2 8 Rxe7 Rxd4 2 9 Rxc7 b5 30 Rc6 Kf7 %-%

93

S t art i n g O ut : A l e k h i n e ' s D e f e n c e

Statistics Not as b a d as you would think from the press, actually. B lack scores

45% after 4 . Nc6, but that i ncludes transpositions to the Exchange . .

variation. On the other hand, when White goes i n for the 'refutation' c4 N b 6 6 e6 fxe6, Bl ack's perce ntage rises to 4 8%.

Theoretica l ? A s theoretical a s you m ake it, really. It's not likely White i s going to know the sharp lines unless he knows you are comi ng, and if you know roughly what the ideas are, you'll stand a chance. On the other hand there is no doubt that preparation i s going to help a lot; there are plenty of games i n Kaissiber where one move turned a good posi­ tion for Black into a quick defeat.

Theoretica l Status Very risky - a s k a nyone! O n t h e other hand, it's hard to i m agine a bigger nightmare for an unsuspecting oppon e n t. Black is taking his life i n h i s hands, but no-one has yet iden tified a line wh ich gives White a clear advan tage.

94

5

C h a pter Fo u r

Exc h ang e Variation w ith 5 R R

. . .

cxd6

I ntro duction I l l ustrative Ga mes

S t a rt i n g O u t : A l e k h i n e ' s Defe n ce

Introduction 1 e4 N£6 2 e 5 N d 5 3 d4 d 6 4 c4 N b 6 5 cxd6

Obviously White i s not trying to knock Black over with th is line. He is content to a i m for a s m a l l edge and keep a space adva ntage of four ranks to three without the onus of defending th e e5-pawn. Mea n­ while, Black can point at the slight weakness of the d4-pawn. This may seem more theoretical than real a t this stage, but in fact Black's main plans in this line revolve a round coun terplay against this pawn. B lack h a s two popular moves. 5 ... cxd6 is examined here, while 5 ... exd6 i s seen in Chapter 5 . The choice between the two is entirely a matter of taste. Each has scored the average 45 or 46% over many games a n d many years. 5 ... Qxd6 does not lose a t once (6 c5? Qe6+) b u t is j ust a slightly inferior move - sooner or later the queen will have to move again to prevent this problem (after 6 B e 2, for example). Before we move on to the analysis of 5 . . . cxd6, it is worth making two points about move order. First, White should probably play 4 c4 be fore exd6, a s 4 exd6 a llows the interes ting possibility of 4 . . . Qxd6. Sec­ o n d l y, o nce White h a s played 4 c4 h e should not continue with 5 N f3 . Th i s is a mix o f systems. Black's best bet i s to p l a y 5 . g6, 6 . . . Bg7 7 0 0 a n d 8 N c 6 and/or ... Bg4, just a s i f e xd6 cxd6 h a d been inserted. White will eventually find th a t he has to play exd6, and meanwhile he will have deprived himself of the most dangerous options. ­

.

.

...

Illustrative Games Game 2 1 0 Gomes • Soppe

ltau Cup 2001

1 e4 N£6 2 e5 Nd5 3 d4 d6 4 c4 N b 6 5 e x d 6 c x d 6 ( D iagram 1 )

Diagram 1 5

96

. . .

cxd6

Diagram 2 All eyes on d 4

. . .

Exchange Variation with 5 . . .c x d 6

5 . . . cxd6 h as always been regarded a s the man's move, with 5 . . . exd6 its t a m e r poor relation. Just lately, h o w ev e r , 5 . . . cxd6 has come under pressure from one soph isticated move order for White, th e so-ca lled Voron ezh variation. This was highly recommended by Jo h n Emms in h is fine Attacking with e 4 and also by Nigel Davies i n his 200 1 Alekhine bo ok ('The Voronezh is a problem, a big one in fact.') , and seems to have driven some Alekhine players away from 5 ... cxd6 alto­ gether. I don't believe things a re as bad as that, and I try to show why in the notes to Game 25 . Before w e come to the Voronezh, though, let us look at the traditional main line, which shows very clea rly what Black's ideas a re in this variation, and hopefully a lso e n ables us to understand the point o f the trendy Voron ezh. 6 Nc3 g6 7 N£3 Bg7 8 Be2 0-0 9 0-0 Bg4 10 Be3 Nc6 (Diagram 2)

This commo n sequence of moves needs no explanation . Both sides de­ velop their pieces to the most n atural places. Black's put pressure on d4, while White's defend it. White has plenty of ways to deviate b e­ sides the Voronezh. I h ave b roken th ese dow n into ideas with h 2-h 3 to stop . . . Bg4 (Game 22), early attacking ideas (a2-a4, h 2-h4, Be3/Qd2, d4 - d 5 and the like - G a m e 23) and Bd3/Nge2 systems (Game 2 4) . M o s t o f these a re more p romisi n g than the natural development seen i n this game, but this sequence is still very popular i n club ch ess. The only re al point to mention, perhaps, is that i f White holds back his queen's k night - say 6 Nf3 g6 7 Be2 Bg7 8 0- 0 0-0 9 Be3 - Black should start with 9 . . . Nc6 rather th an 9 . . . B g4 , which is well met by 1 0 Nbd2, defending the c-pawn a n d nul lifying Black's threat o f . . . Bxf3 and with it much of his counterplay. Right now there a re some tactics to notice - Black is threatening 1 1 . .. Bxf3 12 Bxf3 (obviously 12 gxf3 is u ndesirable) 1 2 . . . Nxc4 . Before this could work Bl ack needed both to castle (so that he didn't d rop the k night on c4 to Qa4+) and to develop the knight to c6 (so that he did n't lose his qu eenside to Bxb 7 ) . This explains White's next move. 11 b3 1 1 d 5 i s n o good a s Black replies 1 l . . . N a5 ! , when White i s really stru gglin g to defend c4 since 12 b3 loses the knight on c3.

l l . .. d5

This i s th e key move fo r Bl ack in this line. 1 1 . . .e 5 is commo n in a ma ­ teur chess a n d i s a lso possible, b u t White re p l i es either 1 2 d 5 or 1 2 dxe5 dxe5 1 3 c 5 a n d sta nds rath er bette r. Nor should Bl ack delay . . . d6-d5 . If he pl ays, for example, l l . . . e6, White can take the bishop pair with 12 h3 Bxf3 13 Bxf3, a n d th en meet 13 . . . d5 with 1 4 cxd5 a n d if n ecessary ca pturing again o n d 5 to produce a situ ation with mutu al isolated queen's pawns. In this type of position development is every­ thing, and White is slightly ahead in that sph ere a n d a lso h as the two bishops; Bl ack is u ncomfortably pl aced. With the text Black wants to fix the d4-pawn once a n d for all.

97

S t a r t i n g O u t : A l e k h i n e ' s D e fe n c e

12 c5

1 2 cxd5 now wou l d b e pointless a s Black c a n recapture on d 5 with pieces and pressure the isolated d - pa w n . 1 2 . . . Nc8 (Diagram 3)

D i a g ra m 3

Diagram 4

H ea d i n g for f5 . . .

A key defender falls

Notice how a l l Black's moves - yes, even this one - are a i m e d at the d4-pawn. Black's plan is to play .. . e 7-e6, ... Bxf3, . . . Nc8-e 7 - f5 . Once that i s done his pressure o n the d - p aw n i s going to tie Wh ite down considerably. Black m ay or m ay not play . . . b7-b6 or . . . a 7-a6 a s appro­ priate. Eventually he will often break with . . . e 5 , a lthough this needs to be ca refully timed, since it weakens the d-pawn. 13 h 3

To m e this doesn't seem very good, a lthough it i s com mon a n d has been played by s t ron g grandma sters . The capture . . . Bx£3 is part of Black's pla n and I don't see any need to e ncourage it. White could a l so play 1 3 Ne 1 (which should be m et with 13 . . .Bxe2 14 Nxe2 b6 while White is not in a positio n to m eet that with b 3-b4) or various develop­ ing half-moves such as 13 Rc 1 or Qd2, which should be m et with the usual ... e 7-e6, . . . N8e7, ... Bxf3, . . . Nf5 p l a n indicated above. (If White seeks to prevent . . . Nf5 w ith g2-g4, i ncidentally, one cou nter- m easure i s . . . f7-f5 , m eeting gxf5 with . . . gxf5 and play down the g-file .) White's m a i n a lternative, however is 13 b4. Th e rash 1 3 . .. Nxb4? i s n o w m et by 1 4 Rb 1 N a 6 a n d then n o t 15 Rxb7? Nxc5! , but 1 5 Bxa6 and 16 Q a 4, when White is m uch better. Instead Black usually plays 13 . . . a6. The idea o f this i s that 1 4 a4? now sim ply loses the b-pawn ( 14 ... Nxb4 15 Rb1 a 5 ) so in o rd e r to go on with his queenside advance White needs to play 1 4 Rb1 , when Black hopes to use the soon -to-be· opened a-file for him self. TIP: When White pushes to b4 in such a position, the reply ... a7·a6 very often holds him up for m uch more than a single move.

98

E x c h a nge V a r i a t i o n w i t h 5 . c x d 6 . .

Black can also play 13 ... Bxf3 14 Bxf3 e6 (when his idea is to find a more suitable moment for . a7-a6, p e rha p s when a 2 - a4 in re ply is im­ possible for so m e reason), but no t 1 3 . e 6, w he n 1 4 b5 Na5 15 Nd2 i s rather a n n oying for t h e a5-knight. In a ny of these cases White wil l advance h is pawn to b 5 , which Black meets with ... Na 5 . White's strong-looking queenside pawns are then surprisingly impotent, a n d Black can aim for p i e c e play o n the kingside or consider breaking with . . . e5. He can also contemplate a well-timed . . . b 7 -b6. O ften c5-c6 i n re­ ply to th is is not to be feared a s a knight can come to d6, where it ha s su perb prospects, while the c -pawn is reasonably easily blockaded. .

.

. .

1 3 ... Bx£3 1 4 Bxf3 e6 15 Qd2 N8e7 16 Ne2

White i s already struggling a l i t t l e with d 4 requiring attention. 1 6 N b 5 N f5 1 7 Bg4 a 6 1 8 Bxf5 axb5 1 9 B c 2 occurred i n a Fischer game, and now 19 ... b4 i s excellent for Black because White has weaknesses on a 2 a n d d4 which he cannot hope to defend forever. ,

16

...

Nf5 1 7 R a d 1 Nh4 1 8 Bg4 h5 1 9 Bg5

Of course White cannot conte m pl a te 1 9 Bf3 Nxf3+. 19 . Bf6 20 Bxh4 Bxh4 (D iagram 4) ..

Thi s exchange is positionally very useful for Black, especially if, as here, he can m a na ge it w i th o u t c h anging t h e p aw n structure b y ca p turing on e 3 . White's dark- squared bishop w as a fine exa mple of Suba's q u i p that bad bishops defend goo d pawns . For exam ple, if the major pieces were now removed from the board W h i te would be help­ less in the ending as h is d4-pawn sim ply c a n n o t be defended . ­

2 1 Bf3 Bf6 2 2 b 4 a 6 2 3 a 4 b 5

A strong move, i llustrating a point I m ade above: White's passed c ­ p a w n is no t to b e feared too much. 24 axb5 Black w a s better anyway, but I feel W h i t e should h a v e t r i e d to avoid this. Once the a - file is opened the battle for the file is only going to see one winner, h a mpered as White is by the back ward queenside pawns being fixed on dark squares.

24 ... axb5 25 R a 1 Qc7 26 g4 hxg4 27 hxg4 Ra7 2 8 Kg2 Rfa 8 29 Rxa7 Rxa7 3 0 Rbl R a 3 3 1 Rb2 Bg7 32 R b l Qd8 Threatening . . Qf6, forki ng bishop a n d d- pawn. .

3 3 Qc1 Qa8 34 Qd2 Qb8! (Diagram 5)

With a threat White can not prevent. 35 Qc1 Rd3 3 6 Qc2 Rxf3 37 Kx£3 Qh2 38 Rd1 e5 (Diagram 6) 3 9 N g3

O r 39 dxe5+ Nxe5 40 Ke3 Qg2 4 1 Kd2 Nc4+ 42 Kc l Bb2+ 43 Kb1 N a 3 + etc. 39 ... e4+ 4 0 Ke3 Qg2 4 1 Nxe4

Otherwise 4 1 . . . Qf3+ wins (4 1 Qe2? Bh 6+) .

99

Start i ng O ut: A lekhi ne's Defe nce

4 1 ...Qxg4

Diagram

5

A key switch

Diagram

6

A timely thematic break

4l...dxe4? 42 Qxe4 puts White right back in the game, but now Black wins the knight in more favourable circumstances, since if it moves away d 4 falls.

4 2 £3 Qh4 43 Qg2 dxe4 4 4 fxe 4 Bxd4+ 4 5 Rxd4 Qe1 + 4 6 Kf4 Nxd4 47 Q g4 Qd2+ 48 Ke5 Nc6+ 0-1 4 9 Kf6 Qd8 i s mate. A one-sided game - Black was much the stronger player - but thematic a n d i n structive nonetheless.

Statistics Black h a s scored 6 0 % from the position after move 1 0. White should definitely seek to deviate fro m this 'natural' method against 5 . cxd6. . .

Theoretical? Not really. All Black's moves are aimed at the d 4 -pawn; a s long a s Black remem bers t o gang u p o n that, he'll b e fine.

Theoretic a l Status Excellent: White has virtually stopped playing like this at high levels. Game 22 0 K hmelnitsky • Palatnik Kherson 1989

1 e4 Nf6 2 e 5 Nd5 3 d4 d 6 4 c4 N b 6 5 exd6 cxd6 6 Nc3 g6 7 h 3 I n t h e last game White's main enemy was the bishop o n g4. This at­ tacked both d4 (by threatening to remove the f3 knight and leave d4 undefended) and c4 (by threatening to decoy the e2-bishop and thus leave c4 undefended). With 7 h3 White resolves to prevent the arrival

100

Exchange Variation with 5 cxd 6 ...

of this turbulent priest. 7...Bg 7 8 N f3 0-0 9 Be2 This is s omething of a key moment. To u n ders tand Wh ite's various ways of arranging his pieces at this point, we need to t h i n k about what Black is trying to do. Sinc e his bis hop cannot now develop to g4 , i t will have to go s omewhere else, presu m a bly f5. T h i s rules out the convenient arrangement of Black's pieces w e s a w in the last game, since h i s knight will not b e able to come from c8 to e7 and f5. I ns tead of this Black normally develops h i s bishop to f5 and his knight to c6, and then arranges the s equence . . . d6-d5, c 4-c5 N c4. For this to be ef­ fective Black needs White to have a bishop on e3 for his knight to hit. So, while White's mos t natural way to de velo p is 9 Be2 Nc6 10 0 - 0 Bf5 1 1 Be3 , this i s playing into Black's hands a little, as we s ee i n the note to White's 1 1th mo ve. White's mos t c unning s equence is the one in the main line of this ga me, but h e does h a ve another idea a t this point, which is 9 Be3 Nc6 10 Rcl (Diagram 7)

Diagram

Diagram

7

White keeps the bishop

on f1

The

8

bi sh o p returns to base

N ow White is ready to meet 1 O . . d5 1 1 c5 Nc4 by taking on c4 in one go, having s aved the tempo spent o n Be2, while h e can meet 1 0 ... Bf5? with 1 1 d5, since l l . . . Na 5 1 2 b 3 is fine with the k night on c3 already defen ded. .

TIP: Once White defe n ds his c3-kn ight he is threaten i n g d4-d5 So lon g as Black remembers that, however, he has two reasonable possibilities. The first is 10

d5 1 1 c5 Nc4 1 2 Bxc4 dxc4 anyway. White scarcely has better than 1 3 0 -0 B f5, and now White has a choice of two fairly forcing lines. This position is analysed in Nigel Davies' book - t h e two critical lines are 1 4 b3 Bd3 15 Rel N xd4 1 6 N xd4 Bxd4 1 7 bxc4 Bxe3 1 8 Rxe3 Bxc 4 1 9 Qxd8 Rfxd8 2 0 Rxe7 b5!, or 14 Qa4 e5! (and not 1 4 . . . Bd3 as White's last introduced the s neaky motif of 15 Rfd l N xd4?? 16 Nxd4 Bxd4 1 7 Rxd 3 !) 15 Rfd l exd4 1 6 ...

101

Starting Out: A lekhine's Defence

Nxd4 Nxd4 1 7 Bxd4 Bxd4 18 Qxc4 Bxf2+ 1 9 Kxf2 Qg5, both o f which are equal according to Davies. Alternatively, Black can change course and, returning to the diagram position, go for 10 . . . e5. This is b ased upon the point that 1 1 d5?! is not so good with White's kingside sleeping- Black can reply 1l. ..Nd4 12 Nxd4 exd4 1 3 Bxd4 Bh6 1 4 Ra 1 Re8+ 1 5 Be2 Nxc4. S o White i s re­ duced to 1 1 dxe5, and now Black can play either 1 l. . .Nxe5 1 2 Nxe5 Bxe5 with the idea of . . . Be6 and . . . d 6-d5, equalising, or 1 l . . .dxe5 12 c5 N d7 13 Bc4 Nd4 with complications which have b een doing quite well for Blac k .

9 Nc6 ...

Black conti nues o n his course, b ut i t has to b e said that right now Black hasn't dem onstrated a good answer to White's main try in this variation. For that reason Davies suggests - and I agree with him­ that Black may i n the fu ture want to investigate a different approach with Movsessia n's 9 . . . e5 10 Bg5 (this is usually the answer to . . . e7-e5 in this line, but may not b e the b est idea here) 10 . . .f6 1 1 Be3 Be6! 1 2 d 5 B f7 1 3 0 - 0 Na6. The bishop is quite good o n f7, where i t helps pre­ vent a c4-c5 break by attacking the d - p a w n , a n d may one day come to h5 after . .f6-f5 and ...g6-g5, and by playing . . . e 7 -e5 early Black has been able to develop h is queen's knight to the useful a6-sq uare in­ stead of having it kicked b ack fro m c6 to e 7 after d 4-d5. .

1 0 0-0 Bf5 11 Bg5 This was Whi te's latest idea before the Voronezh w as inv ented and, theoretically, i t still sta nds very well. 1 1 Be3 is much less pointed: Black replies 1 l. . . d 5 1 2 c5 Nc4 1 3 Bxc4 dxc4 and achieves the position a nalysed in the note to White's ninth, but without White's Rc 1 , which m ust favour Black slightly. 1 1 Bf4 is usually the s a m e thing - see the next note.

ll h6 ...

1l...d5? 1 2 c 5 N c 4 is pointless since White can just chase i t away with 13 b3. Funnily enough, 1 l . . . h 6 is also t h e main reply to 1 1 Bf4. Black is now threatening . . . e7 -e5 (an immediate 1 l . . . e 5 is supposed to b e strongly met by 1 2 Bg5, a lthough it's n o t a t all obvious to me that Black is b adly placed after 12 . . . Qc8). White can play various moves after 1 L..h6, such as 1 2 Rc1, which allows 12 . . . e5 when Black is fi ne, or 1 2 Qd2, when Black plays 12 g5 ( 1 2 . . . Kh7 is too slow; once White has protected the c3-knight h e is ready for 13 d5! which gives him the advantage) and follows with . . . d6-d5 or . . . e7- e5 as a p propriate. White can also venture 12 d5 Na5 1 3 N d 4 , tossing a pawn to loosen Black's structure, a fter which 13 . . . Naxc4 14 Nxf5 gxf5 15 b3 Ne5 16 Qd2 Kh7 17 Rac 1 Ng6 18 Be3 Rc8 is a natural seq uence that gives Black good chan ces. But the m a i n l i ne is the paradoxical 1 2 Be3, transposing to the game. The point is that . . . e7-e5 is much more powerful when it at­ tacks the b ishop on f4 (or shuts i t out of play after 12 Bg3 e 5 13 d5?!) since White cannot meet it with a n i m mediate d4-d5. . ..

102

Exchange Varia tion with 5 ...cxd 6

You would think that since Blac k does not really want to play . . . d6-d5, c 4-c 5 N c 4 while W h i te still has his bishop o n c 1 , i s s u pposed not to be threatening ... e7-e5 until h e plays ... h7-h6 because of Bg5, and has run out of natural developing moves to play before then, that White c ould c ome up with s ome developing move which didn't involve this Bc1-f4 -e3 pendulu m . 1 1 Re1, for example, looks like a pretty s ens i ble move, although as far as I know it hasn't been played. 1 1 b 3, other hand, a llows Blac k to change course e ffec ti vely with 1l.. .d5 12 c 5 N d 7 when the pres s ure on t h e long diagonal is inconvenient, for example 13 Bb2 Be4 1 4 Na4 e5!.

1 2 Be3 Not 12 Bf4? e5 etc.

12 . . .d5 13 c5 13 b3 dxc4 14 bxc4 Rc8 15 Rc1 Na5 1 6 c 5 N6c 4 is fine for Blac k.

1 3 . . .Nc4 14 B e l ! (Diagram 8) This i s the idea. 14 Bxc4 dxc4 leads to variations simi lar to those a n a­ lysed above i n the long note to White's n i n t h . Black is fine there, but now h e appears to be rather s nookered. How i s the threat of 15 b3 to be met?

14 ... b 6 15 b3 bxc5!? This m ay in the end be unsound, but White has n't quite proved this. The a l ternative i s to retreat meekly with 1 5 . . . N4a5 16 c xb6 axb6. If Black has to exile his k n ight to a5 and allow an outpost o n b5 it looks like his play has not worked out, and i ndeed practice favours White cons iderably from this position, although i t is ce rtainly playable for Black a n d h e m a y one day be able to elaborate plans which will see him all ri ght, although i t hasn't happened yet.

16 bxc4 cxd 4 17 Nxd5 e6 The other try is 17 ... d3 18 Bxd3 Bxd3 19 Qxd3 Bxa 1 2 0 Bxh6 (beli eve it or not, this is the whole point of Whi te's bishop s huffle a t moves 1 1, 12 a n d 14) 20 . . . Bg7 2 1 Bxg7 Kxg7 22 Qc3+. Now 22 . . .£6? loses bac k the exchange to 23 Ng5 Qd6 24 Nc7, so Black c ontinues 22 ... Kg8 2 3 R e1 and n o w White is supposed to b e virtually winning w i t h h i s ac­ tive pieces and ability to create threats against the king. It see m s to m e that 2 3 . . . Rc8 still gives Black some hope of defending, but this position i s surprisingly u n attractive for h i m.

18 B a 3 (Diagram 9) A novelty when this gam e was played; earlier games had featured the limp 18 Bb2 . 18...R e 8 18 . . . exd5 1 9 Bxf 8 s i mply wins materi al.

19 N f4 e5 A slightly fortunate tactic wins Black back his piece, but he remains under pres sure.

103

S tarting Out: Alekhine's Defence

20 Nd5 d 3 2 1 Bxd 3 Bxd3 22 Qxd3 e4 2 3 Qe3 exf3 2 4 Qxf3 N e 5 24 . . . B x a 1 25 R x a 1 is too dan gerous.

25 Q b 3 Rb8 26 Q a 4 Rbc8 27 Rac1 Qh4 28 R fe 1 Re6 29 Qxa7 Rxc4 30 Ne7+ Kh7 3 1 Qb8?? (Diagram 1 0)

Diagra m

Diagram

9

T he bishop come s alive again

10

Is the party over for Black?

Both sides were in time trouble. The text loses to 3 1 . . .Rxe7 32 Bxe7 Rxc 1 33 Bxh4? (33 Rxcl Qxe7 is forced, but Black is better) 33 . . . Rxe l + 3 4 Kh2 Nf3+ a n d 35 ... Be5+ spearing the q u e en; correct w a s 3 1 Qa8.

3 1...h5? 3 2 Qg8+ Kh 6 3 3 Re3 It seems that 33 Rxe5, with the cute twin points of 3 3 . . . Bxe5 3 4 Nf5+

gxf5 35 Bf8+ ( m a ti ng) and 33 . . . Rxe5 34 Qxf7 (forking c4 a n d g6) 34 ...Qxe 7 35 Qxe7!, would have won- aren't comp uters clever?

33 ... Rxe7 3 4 Rxc4 Qxc4 3 5 Bxe7 1 -0 (time) Black lost on time j us t when his courageous defence had attain ed a drawn position ... after 35 . .. Qc 1 + 36 Kh2 Qc7 37 f4 Qxe7 38 fxe5 Bxe5+ 39 Kh 1 Qc5 40 Re 1 Qf2 Black can force perpetual check on the dark squares around the king.

Statistics White has scored 5 3% from the position after 1 0 ... Bf5.

Theoreti cal? Fairly - Black does need to be aware of the main l i ne and the tactics in the notes to White's ninth move. O n the other hand the m a i n line is practically forced from Black's point of view, and thus quite easy to remember.

1 04

Exchange Varia tion with 5 cxd 6 ...

Theoreti cal Statu s N eeds work, t o be frank. If I w a s White I'd forget the Voronezh a n d go bac k to this. Black m ight have to look into 15 . . . N 4a5. Game 23 0 Salmensuu • Solozhenkin Finland 2000

1 e 4 Nf6 2 e 5 N d 5 3 d 4 d 6 4 c4 Nb6 5 exd6 cxd 6 6 N c 3 White h a s two alternatives a t t h i s p o i n t . The first i s 6 d 5, which h a d a brief vogue b u t really has no merit unless Blac k continues o n autopi­ lot with 6 ... g6? 7 Qd4 ! . Black's simplest reply i s 6 . . . e5. The other i s something more or l e s s insane, starting with 6 Be3 g6 a n d n o w either 7 h4!? or 7 Qd2 followed by Bh6 and h2 - h 4 etc . Either way Black does best not to be afraid, but to play 7 . .. Bg7 and c astle. H e can strike back in the centre with e7 -e 5, and meet h5xg6 with ...fxg6 if necessary i n order to defe nd h7 along the second ran k . The positions are compli­ c ated and little explored, but this lack of ex ploration is bec ause Black obvi ously has dangerous c hances of his own. ...

6 ... g6 7 a4 White intends either to gain space on the queen side with a 4 -a5, R a 3 and d4-d5, or to e s s a y the c lever tactical sequence i n t h e g a m e . White could have pushed the a-pawn o n move 6, which i s still best met with 6 ... a 5 .

7 . a5 ..

Black can also play 7 . . . Bg7 8 a5 N 6d7, but this i s rather harder to handle and hasn't been proven equal. 8 c5!? (Di agram 1 1 )

Diagram

11

White e xerts pressure

Diagram

12

A clever defence

This is White's idea. N igel Davies had some success for a while with

105

Starting Out: Alekhine's Defence

the quiet 8 Nf3, but Black can deal with this by simply developing, for example 8 ... Bg7 9 Be3 0 - 0 1 0 Be2 N c6 1 1 d5 (White has prevented l l ... Na5, but alas . . . ) l l . . . Ne5 12 N xe5 Bxe5 13 0-0 Nd7. This is the prob l e m . This knight i s very well placed on c5 and can no longer be removed by b 2 -b4. Black has equali zed. However, 8 c5 is a c h a ll e n gi n g effort whi ch Black needs to take seri­ ously.

8 ... dxc5 9 Bb5+ Of course 9 dxc5 Qx dl+ is fee ble, but now Black has to take evasiv e acti on si nce 9 . .. N 8d7?? 1 0 dxc5 and 9 . . . N c6?? 1 0 d 5lose a pi ece, a n d 9 . .. Bd 7 1 0 dxc5 is a l s o ugly. Note t h a t t h e reason White wanted to throw in 7 a4 a5 before 8 c5 was so that the bishop check cannot be met by . . . N c6, d 4 - d 5 a 7 - a 6 etc.

9 ... N 8 d 7 1 0 B£4 Of course White can still play 10 dxc5 but the sti ng has been taken out of it now that the knight has left b6; Black simply develops with ... Bg7, . . . 0-0, . . . N c6 and is fine. I n fact . .. Bg7 is also the answer to moves such as 10 Nf3 or 1 0 Bg5, although in this latter case Black needs to be happy about 10 . .. Bg7 1 1 N d5, whi ch seems to force 1 l. . .f6, although this looks okay to me. Alternatively h e could first play 1 0 . . . h6. I O Bg7 •..

11

Nd5

The culmination of Whi te's tactical sequence. He now threatens N c7+.

ll...N a6 (D iagram 1 2) This was played for the o n e a n d only time in this game, but seems to present a complete answer to White's i dea. The old line was 1 l ...e5 1 2 dxe5 0-0 1 3 Nf3 N c6 1 4 Bg5 (or 1 4 Bxc6 a n d 1 5 Bg5) 1 4 . . .f6 (which is sti ll given as t h e answer to 8 c5 in most books) but White has been do­ ing rather well there. 1 1 . . . 0- 0 loses the exchange or so after 12 Bc7 Qe8 1 3 Nb6 Ra6, although Black certainly obtains some compensa­ tion. 12 dxc5

At first sight 12 Bxa6 wi ns material, but Black has 12 . . . e 5 up his sleeve. 13 Bg5? Qxg5 1 4 N c7+ Kd8 1 5 N xa8 Qxg2 16 Qf3 Qxf3 1 7 Nxf3 bxa6 t h e n wi ns for Black, who will round u p the knight, while 13 Bxe5 N xe5 14 Bb5+ Nc6 is j ust good f or Black. Mter 13 dxe5 Black recaptures on a6 with the rook (and meets 14 e6?? with the unkind 14 .. . Rxe6check). The critical line may be 12 Bxa6 e5 1 3 Bb5 exf4 1 4 Qe2+ Kf8 1 5 Nf3, but I don't thi nk White h a s much t o be proud about after 15 . . . Bxd4 16 N xd4 cxd4 17 0-0 Qg5 18 Nc7 N e 5, as in Ansel l­ Cox, London League 2004, and probably still less so after 1 5 ... cxd4 1 6 0-0 Nf6 1 7 Nxf4 Qd6 fol lowed by . . . Bh6 and ... Kg7. Nonetheless this was surely better than the ga me, in which White qui ckly finds him­ self unable to def end the c5-pawn

1 2...0-0 13 Re i e6 14 Nc3 e5 1 5 Be3 Nd xc5 16 Nge2 Be6 17 0-0 N b 3 1 8 Q x d 8 Rfxd8 1 9 Rcdl N d 4 20 Bg5

1 06

Exchange Variation with 5 . .cxd 6 .

White has nothing for his pawn and the Russian gra ndmaster c annot have been too thrilled with his technique i n what follows.

20 ... £6 21 B e 3 f5 2 2 Bg5 Rdc8 2 3 Nxd4 exd4 24 Ne2 Bc 4?! 24 ... h 6 (removing the bishop from the h4-d8 diagonal) fol lowed by .. . Rd8 was better.

25 Bxc4 Rxc4 26 b3 R b 4 2 7 Be 7 d 3 I expect you w a n t t o know why 2 7 . . . Rxb3 was not played, but I'm afraid I can't help you. Black has some slightly irritating problems in coordinating his pieces, but it's hard to believe h e isn't still be tter. Now the game peters out.

28 Nc 1 R e 4 2 9 B a 3 N b 4 3 0 N x d 3 Nc2 3 1 Bb 2 Bxb 2 3 2 N x b 2 'h-'h

Statisti cs White has scored trem endously after 7 a4, but not quite so well after 7 ... a 5, a n d in any event there is only one game with l l . . .N a 6 ! . This i s largely because t h e popularity o f t h e line h a s p l u m meted since this reply was discovered.

Theoretica l ? Not enormously, b u t Black can lose quickly if he doesn't know t h e l i n e u p t o m o v e 1 1 a n d t h e 12 B x a 6 e 5 trick. It's a l s o worth re m e m be ring 6 d5 g6? 7 Qd4.

Theoretica l Statu s 7 a4 c a n tentatively be placed i n the 'solved' basket. I t's always risky to write off a line after one ga m e , but I'll be surprised i f l l . . .Na6 isn't the whole answer. The Dragon- style attac k s with h2-h4 are theoreti­

cally nonsense, but they will always appeal to the basher. Game 24 0 K.Spraggett • Nataf Enghien-les-Bains 1997

1 e 4 N£6 2 e5 Nd5 3 d4 d 6 4 c4 N b 6 5 exd6 cxd6 6 Nc3 g6 7 B d 3 B g 7 8 Nge2 (Diagram 1 3) This introduces a dangerous positional system which was very fash­ ionable in the early 1 9 90's but has since fallen out of vogue with the rise of the Voronezh. It was the choice of the young Bobby Fischer, al­ though later h e changed sides to uphold Black's colours i n the 1970 Interzonal o n h i s w a y to the world championship in 1972. White's idea is t h e s a m e i n a w a y a s Game 2 2 - he wants to take t h e sting out of ... Bg4 wh ich, clearly, c a n now be met with f2-f3 (and . . . Bf5 i s pre­ vented). Moreover the k night o n e2 defends its colleague o n c3 and makes possible b2-b3 or d4-d5 without needing to waste time on Rcl.

107

Starting Out: Alekhine's Defence

Diagram

13

Unfashionable but dangerous

Diagram

14

Pressure on the

centre

White's idea, generally speaking, is to m eet . . . Nc6 with d4·d5. In the structure w hich then arises h e has no objection to his d3-bishop being re moved for a knight. Almost always he will send his bishop to d4 to initiate a trade, after which his knight on d4 will usually be more than a m atch for Black's bishop on d7. Black's problem is that he can­ not allow the pawn structure to remain as i t is- if he does he will be run n ing the risk of being crushed on the kingside- and yet the na tu ­ ral break ... b7-b5 is very hard to arrange with White's knights on c3 and d4, and a break with ... e 7 -e6 w i l l always l eave p a wn weaknesses and a slightly draughty king.

Black, m eanwhile, has the choice of three ways of playing. He can go for ... Nc6 and allow White the type of ga m e describ ed above, h e can go for ... e7-e5 before playing ... Nc6 or h e c an adopt a ... d6-d5 p lan, as Fischer did and as Nataf does in this game. 8

•••

0-0

Or 8 . . . Nc6 9 d5 Ne5 10 0-0 which is likely to transpose below.

9 0-0 White h as an occasionally p l ayed but thematic alternative here in 9 d5, which is designed to achieve the pawn structure described above whil e cu tting out som e of Black's other ideas (such as that played in this gam e). Black coul d accept this by 9 ... N8d7 followed by Ne5, or he could play 9 . .. e5 (10 dxe6 Bxe6 is feeble) but in my opinion his most logical option is 9 ... e6, followed by ... exd 5, . . .N8d7-e5, . .. Re8 and so forth . I have to admit that as yet no Black player h as seen it m y way, though . Slower moves like 9 b3 or 9 a 3 can also be m et with . . . d6-d5, w hich is not so good for White.

9 ... d5 If this does not a p peal Black can a lso play 9 . . .Nc6, although in m y opin­ ion White is som ewhat better after 10 d5 Ne5 11 b 3 Nxd3 12 Qxd3 Nd7

108

Exchange Variation with S...cxd 6

1 3 Be3 Nf6 (this is the best square for the knight, from which it assists both the ... b7-b5 and ... e7-e6 breaks indirectly, and prevents an imme­ diate exchange of da rk - s qua red bishops, leaving White a little uncer­ tain whether bishop or knight should occupy d4). If Black does want tD play this way, the plan . . . Bd7, . .. a7-a6, . . . Qa5, ... Rfc8, as i n !.Rogers­ Moor, Baden 1 999, where Black gave a way 300 rating points and still drew handily, is probably best, giving White the problem of holding down . . . b7-b 5 while organizing h i s forces on the ki n g side . Alternatively, a long-established l i n e is 9 . .. e5 10 Be3 ( 1 0 d 5 , enabling White to preserve the d3-bishop, i s also possible and perhaps quite good) 10 . . . Nc6 (or perhaps 10 ... f5!?) 1 1 d 5 Nb4 12 b 3 Nxd3 13 Qxd3 f5 (or 1 3 . . . Re8) 1 4 f4 N d7, a n d now Black h a s t o be prepared to meet ei­ ther 1 5 Rad1 or 1 5 Qd2 with 1 5 . . . g 5!?, leading to extremely sharp play o n the kingside, if h e wants to avoid ge ttin g positionally crushed. Otherwise after . . . e5-e4 White puts his bishop on d 4 and a k n i g h t on e3 and advances on the queenside, or i f Black maintains the tension White will a t some point take on e5 and (if Black recaptures with a piece) go for Bd4xg7 and then N d4, or if Black recaptures with a pawn, aim to queen t h e 'd' and c-pawns. Black's queen's bishop is a poor piece in these lines. These positions have more i n com mon with the King's Indian than with other variations of the Alekhine.

1 0 c 5 N6d 7 11 Bg5 h6 12 Be3 Nc6 (Diagram 1 4) 13 Bb5 A strange-looking move, but Black w a s threatening to liquidate ad­ vantageously with . . . Nxc5 and ... d 5 - d4. 1 3 Nxd5 N xc5 1 4 N xe7+ Qxe7 15 dxc5 Bxb2 16 Rbl Bg7 17 Qb3 Rd8leads nowhere i n particular, while 16 Bxh6 Rd8 1 7 Rbl Be5 1 8 Be3 Bf5 gives Black fine compensa­ tion for the pawn

13 ... Nf6 14 f3 14 . . . Ng4 was threatened.

1 4 ... B f5 1 5 Qd 2 Kh 7 1 6 N g 3 Bd 7 1 7 Rab 1 Na5 18 b3 b 6 1 9 Rfd 1 Perhaps White needed to swallow h i s pride and take on b6, but if he does then clearly his opening h a s been a failure.

19 ... Bxb5 2 0 Nxb5 bxc5 21 d xc5 R b 8 2 2 Nc3 e6 23 Nge2 Nd7 24 f4 Qc7 2 5 Rd c 1 Rfe8 The best move i n the g a m e , which is designed to prevent exactly what Spraggett now plays.

2 6 Nd4? (D iagram 15) 26 .. . Qxc5 27 Ncb5 The point of Black's 25th was that 27 Nxe6? now loses to 27 . . . Qe7,

pinning the knight to the bishop on e3.

27 ...Q b 6 2 8 N c 7 Rec8 29 Ndxe6 d 4 (D iagram 1 6) White has been out-combined. Despite the out-of- play knight on a 5 h e has no w a y both t o rescue h i s over-com mitted pieces and coordinate the ones he m a nages to ret ai n.

109

Starting O u t: A lekhine's Defence

Diagra m

Diagra m

15

White is happy to see

.

.

.

Qxc5

16

The complications favour Black

30 Nxg7 d x e 3 3 1 Q x d 7 Kxg7 32 Kfl Or 32 Nd5 Rxc1 + 33 Rxc 1 Qd4.

3 2 .. . R d 8 33 Q e 7 R d 2 34 Q e 5+ K g 8 35 b 4 Q b 7 36 N d 5 R x d 5 0-1

Stati stics Worrying. W h i t e h a s scored 62% a nd t h i s rises t o 66% in the l ines with 9 . . . N c6. With 9 . . . e5 Black has done much better, scoring 45%, and with 9 . . d 5 Black has a considerable plus score, but from a smal l sample. .

Theoreti cal? If you want to play an . . . e7 -e5 or ... Nc6 p l a n then knowing the correct placement of the pieces and what White is trying to do is m or e impor­ tant than specific moves. The . d 5 plan is tr ickier and at l east a n idea o f the tactics hanging around at move 1 3 or so i s vital . . .

Theoretica l Statu s Not many games, but White hasn't got to grips with this 9 d5 plan yet in the way he has with the older l ines. 9 . Nc6 gives White a stable ad­ vantage and this doesn't seem set to change. 9 e5 10 Be3 has been considered okay for Black for some time, although he needs to be will­ ing to accept very sharp play. 1 0 d 5 would trouble me more in this l ine. ...

.

.

...

Game 25 0 D.Coleman • Hillarp.Persson Copenhagen 1999

1 e 4 N f6 2 e 5 N d 5 3 d 4 d 6 4 c4 N b 6 5 e x d 6 cxd6 6 N c 3 g6 7 B e 3 B g 7 8 R e i 0-0 9 b 3 (D i a gram 1 7)

110

Exchange Variation with 5 .. .cxd 6

T h i s is it. The dreaded Voronezh vari ation. Voronezh, by the way, i s a town i n Si beri a. The line was invented b y the citizens of this fair burg - and in particul ar worl d c orre s p onden ce champion Grigory Sanakoev -i n the mid-to-late 1980's, and gradually seeped out i nto t h e West around the mid - 1 990's. Ho pefully, after the previ ous games we can now see Whi te's idea. He is very wel l prepared for 9 . . .N c6. Just as i n the l ast game he wi l l meet that with 10 d5 N e 5 but this time there isn't even a bishop on d3 for the kn ight to be traded for, a n d White will fol l ow u p i n due course with f2-f4, driving the poor beast back to d7, where it prevents Bl ack from developing his queen's bishop or recycling the b6-kni ght to somewhere useful. This has scored terri bly for Black. Meanwhile Whi te has also prevented 9 ...d5, which i s now senseless- 1 0 c 5 is al­ re ad y winning material. N ote that 9 ...d5 10 c5 N c 4 would have been the answer to 9 Be2.

Diagram

17

The Voronezh

Diagram

18

An important t heoretical position

Black h a s toyed with 9 . . . f5 ! ? 10 g3 and 9 ... Bf5, whi l e Davi es suggests 9 . . . e6 and/or 9 .. . Bd7, but the next is the main line.

9 ... e 5 D a vies i s wonderful ly s niffy about this move- 'It coul d be that some­ one l i k e Baburin wi l l eventually show how Bl ack can draw' - but I thi nk Black has decent enough play.

1 0 d x e 5 d xe5 1 1 Qxd8 White c a n also play 1 1 c5 N 6d7 12 Bc4 when 12. . .N c6 13 N f3 Qa5 l eads to a raft of compl i cated o l d theory, probabl y quite good for White, but Baburi n's excel lent 12 ... Qa5 seems to have put t h is line out o f busi ness. The point is that 1 3 N f3 i s not good because o f 1 3 ... e 4 1 4 N d4 N xc5, so White h a s t o devel op w i t h 1 3 N e 2 w h e n 1 3 . . . Nc6 (and not 13 .. . Nxc5? 1 4 Qd5, which l eaves Bl ack i n troubl e) 14 a3 Nd4 15 Nxd4 exd4 16 Bxd4 Re8+ 1 7 Kfl Bxd4 18 Qxd4 Qxa3, while sti l l

111

Starting Out: Alekhine's Defence

very compli cated, seems to be good for Black

1 l . .. Rxd8 1 2 c5 This i s White's prepared way to meet 9 ... e5. He drives the knight back to d7, where it blocks most o f Black's pieces, si nce 12 . Nd5? loses to 13 Rdl Be6 14 Bc4. 12 Nb5 is refuted by 12 . . . Nc6 13 Nxa7 Nd4! etc. ..

12 N6d7 13 Bc4 Nc6 1 4 N f3 (Di agram 1 8) ...

1 4 Ne4 is also supposed to be slightly better for White after 1 4 ... Nf8 1 5 Nd6 (or 15 Bg5 Rd7 1 6 Nf6+ Bxf6 1 7 Bxf6 Rc7 1 8 Nf3 Bf5) 1 5 . . . Ne6 1 6 Nf3 Ncd4 1 7 Ng5 Nxg5 1 8 Bxg5 Rd7, but a s long as Black conti n­ ues wi th the key ma noeuvre ... Rc7, . . . Be6 I don't thi nk he has too m uch to fear. The di agram position i s key to the variation and ari ses most of the time after 9 . . . e5. Whi te's active pieces and pressure against f7 are evident. He also has the p rospect of Ne4-d6 andlor b3-b4 to cement his pawn wedge on the queensi de, followed by pushi ng Black backwards over the edge of the board. Black needs to sh ift the d7 - k ni ght. The most natural plan, perhaps, is to send it to f8 and e6 to target d4, but many games have featured thi s plan pi tted against White's ideas described above, and the verdict is unequivocal - Black is too slow. Instead he needs to find a pl a n which both breaks White's queenside formati on and frees hi s knight from d7, and happily there is one available.

1 4 ... Na5 An essential link i n the plan. The i mmediate 1 4 ... b6 allows 15 Bd5 Bb7 16 Ne4, which i s extremely u n p leasant. 15 Be2 The original idea of the Voronezh group was 15 Bg5, w h ich was what Sanakoev played when he first i ntroduced this l ine. However White has got nowhere against 15 ... Rf8. White's i dea i s 16 Be7 a nd now rather than 16 ... Re8 1 7 Nd5 with an i niti ative for Whi te, Black should shatter Whi te's pawns with 16 . . . Nxc4. According to Sanakoev White can sti l l retai n a sli ght advantage after 17 bxc4, although I don't see why after 1 7 . . . Re8 1 8 N d 5 e4 1 9 Ng5 b 6 . In practice White has gen erally grabbed the excha nge with 1 7 Bxf8 Kxf8 18 bxc4 Nxc5 19 0-0 e4 20 Ng5 Nd3, but White can a t best draw here (2 1 Rc2 Nb4 22 Reel Nd3 and so o n) whi l e Black can deci de whether to play on (wi th 2 l . . . f5, for example).

1 5 ... b6 White i s meant to be better here, but I believe this i s the answer to the Voronezh.

16 cxb6 Usually 1 6 N a 4 is reco m m ended. 'rhis is based on a game where Black i ncauti ously played 16 ... bxc5, no doubt expecti ng the routine 1 7 Nxc5, a n d was i nstead met b y 1 7 0-0 ! . Black could i n fact have avoi ded thi s position with 16 . . . Nxc5 1 7 Nxc5 bxc5 18 Rxc5 Nb 7 when he is fine, the best plan against most replies bei ng to send the b 7-

112

Exchange Variation with 5 . . .cxd6

knight via d6 to f5. White in his turn, though, can force it with 1 6 0-0, w he n Black hardly has better than 16 ... bxc5 17 Na4. But anyway 17 . Nc6! is fine for Black, e.g. 18 Nxc5 Nxc 5 19 Bxc5 Bb7. White c a n also p l a y t h e sl ightl y al arming-looking 16 b4 Nc6 17 b5 Nd4 1 8 c6 Nc5 but Black can break up t h e pawn mass with ... a 7-a6 and equal ise; 1 9 Bc4 Be6 2 0 Bxe6? Nd3+ i s a n important point. ..

16 ... Nxb6 1 7 N b 5 Bb 7 1 8 0-0 1 8 Nxa7 Nd5 gives Black good pl ay

.

1 8 ... Nd5 1 9 Bg5 19 R fd 1 N xe3 20 fx e3 e4 2 1 Nfd4 Bf8 is fine too, while 1 9 Bxa7? Nf4 i s al ready very good for Bl ack.

19 ... R d 7 20 R fd 1 a6 2 1 N a 3 T h e problem with White retaining his bishop with 1 9 Bg5 is that now 2 1 Nc3? l oses a piece to 2 l . . . Nxc3 22 Rxc3 Rxdl + 23 Bxd1 e4, so the k n i g h t i s forced to this poor square.

2 1 .. . e4 22 N e 1?? 22 N d4 Nb4 was al ready quite unpleasant for White but the text l oses material to a neat sequence.

22 ... Bb 2 23 Rc5 Bxa3 24 Rxa5 Bb4 2 5 Ra4 Bxe 1 26 Rxe 1 Nc3 2 7 Bg4 Rd 5 2 8 Bf6 N x a 4 2 9 bxa4 Rd 2 3 0 R c 1 e3 0 - 1

Statistics Black h a s scored 3 6 % in t h e Voronezh a s a whole and 42% after 9 . .. e 5 , but I think this reflect s t h e fact B l a c k players haven't yet g o t round t o the right pl an. If y o u don't bel ieve m e , according to m y Barbican team- mate Dave Coleman, Tiger Hil l arp-Persson was extremely con­ fident about Black's position i n the post mortem.

Theoretica l ? I fear s o , yes. Given that i t h a s been recomm ended i n a very popular repertoire book, you are very l ikel y to meet this l ine in practic e. Black reall y needs to know up to move 1 5 of the m a i n l ine. To be fair, the first eight moves are routine, and most of the next few are recap tures, so t hat isn't quite as bad as it sou nds. If l earning by heart isn't your thing, keep in mind that the key idea is to get rid of the d7 -knight a n d break White's p a w n p h a l a n x . It's a l so worth trying t o rem ember Ba­ burin's 12 Qa5. . ..

Theoretical Statu s . cxd6 Ex-

Canonised by pr e vious writers and supposedly has the

..

change Variation in c risis for Black. I don't believe this

. . .

make up

your own mind.

113

Chapter Five

Exchange Vanation . with 5

exd6

...

�

Introduction

�

Illustrative G ames

Exchange Variation with S ...exd6

Introduction 1 e4 Nf6 2 e5 Nd5 3 d 4 d 6 4 c4 N b 6 5 e x d 6 exd6 i ntroduces a di f­ ferent k i n d of game from 5 ... cxd6. Here Black i s c ontent with a sym­ m etrical structure. O n t h e whole, this li ne is more soli d than recap­ turing with the c -pawn but i n so being offers Black less potential for counterplay. Whi te has a very simi lar range of i deas available to the previ ous straightforward development (Game 2 6) , development with a more cunning order (Game 27), a Bd3/Nge2 syste m (Ga m e 28) and vari ous l u n atic attacking i deas (Game 29).

c hapter:

As for Black, there is really only one pawn structure dec ision - whether

to play ... d6-d5. The typical development runs along the lines of ...Bg4/f5 , ... Be7-f6, . ..0-0, . ..ReB and either . . . Nc6 or ... c7-c6.

or not

Illustrative Games Game 26 0 Gipslis • Larsen Sousse 1967

1 e4 Nf6 2 e 5 Nd5 3 d4 d6 4 c4 Nb6 5 exd6 e x d 6 (Dia gram 1 )

Diagram

1

Modest development

Diagram

2

What next?

6 Nc3 Be7 Black would quite like to develop with . . . g7-g6 a n d Nf3 Bg7 8 Bg5 is irri tating.

... Bg7

b u t 6 ... g6 7

7 N f3 0-0 8 Be2 Nc6 9 Be3 Bg4 10 b 3 Black was threatening 10 ... Bxf3 1 1 Bxf3 Nxc4. After 1 0 d 5 B x f3 1 1 Bxf3 Ne5 White i s forced to play 1 2 b3 a n d thus c a nnot retai n his bishop- 12... Bg5 13 Bb2 Nxf3+ 14 Qx£3 Bf6 is e q u al .

115

Starting Out: A lekhine's Defence

1 0 ... Bf6 1 1 0-0 1 1 h 3 Be6 12 Ne4 (or the i m mediate 1 1 Ne4) leads to a similar game where Black is slightly less c ra m p ed but, on the other h a n d , White h as the two bishops.

1 l.. .d 5 Black c a n a l so - and perhaps should - delay this with 1 l . . .Re8 12 Rcl d5, s ince d4-d5 is not possible until c3 is guarded. Having said that the rook c a n also be useful o n the f.fi l e (see the note to White's thir­

teenth) and White's Rc1 is usefu l inasmuch as it discourages .. . b7 -b6, so i t is not clear that this trade-off favours Black ...

1 2 c5 Nc8 (Diagram 2) 1 3 b4 According to Larsen this is a bad move as White is not achieving any­ thing dra matic with b4-b5. It might be different if Black's bishop were

not on g4, where it has the prospect of removing the f3-knight and drawing t h e bishop fro m its defence of c4. Instead many games have continued 13 h3 Be6 (best as 13 . .. Bxf3 1 4 Bxf3 leaves White better­ h is plan should be to adva nce on the kin gsi d e - and 13 ... Bf5 simply loses a tempo because g2-g4 is usually part of White's plans si nce it is usefu l to prevent . . . Nf5, and a general kingside pawn advance is White's most dangerous a p proach in this line) 1 4 Qd2. Black's best i s either 14 ... N 8e7 or 14 . . . g6. T h e idea o f t h e latter option i s th a t Black is concerned about 14 ... N 8e7 1 5 g4 (t h re a ten i ng g4-g5) 1 5 . . . g6 16 Ne 1 Bg7 17 f 4 f5 1 8 g 5 , gai nin g a great d ea l of space, and now intends, by saving the tempo of ... N8e7, to meet this plan with 14 . . . g6 1 5 g4 Bg7 16 N e 1 f5 when White h a s no t i m e for 1 7 g5 in view of 17. . .f4, and thus has to moderate h i s ambitions on the kingside. These lines are typic al; Black usually has to meet White's kingsi de a dvance either with .. .f7-f5 or . . . h 7-h6 a n d ... g6-g5 in t h i s line.

13...N 8 e 7 14 b5 Na5 No doubt Gipslis h a d played 13 b4 i n order to b e able to play b4-b5 when e 7 was occ upied a n d thus force t h i s k n i g h t here b u t , i f this was the c ase, the piece's future career m u st have disappointed h i m .

1 5 h 3 Bxf3 Of course: now Black can send his k n ight to c4 after the recapture. White's queenside adventure h a s ruled out the normal plan o f a k ing­ side advance. 1 6 Bxf3 c6 ( D ia gram 3) Blac k i s already better according to Larsen . As h e points out, Black has n o weakness, but Wh ite does - d 4 .

1 7 Qd 3 Nc4 18 Bf4 Ng6 1 9 Bh2 Bg5! A fine move. As Larsen explai ns, although his bi shop may be formally better than White's, i t needs to be traded to let t h e Black queen go to f6 and k n ight to f4. Black m i g h t also need to put a rook o n b8 at some point.

2 0 bxc6 bxc6 21 Bd 1 B f4 22 Bc2?

116

Exchange Variation with S ...exd 6

According to Larsen 2 1 Bd 1 was a fi n e mov e wh ose object ought to

h av e been to bring the bishop to b 3 to chase away t h e knight.

D iag ram

Diagram 4

3

The d4-pawn mig ht prove we ak

The d4-pawn falls

2 2... Bxh 2+ 23 Kxh 2 Qf6 24 g3 Rfe8 25 Kg2 By now 25 Bb3 could b e answered with 25 . . . Rad8, but the text ov er­ looks some tactics.

25 ... Qg5 26 Kh2? Th e vile 26 f4 Qe7 was forced.

26... N b 2! 27 Q f3 Qd 2! 2 8 Bxg6 hxg6 (Dia gram 4) And th ere goes the d-pawn. 2 8 Rab 1 Rab8 ch anges noth ing.

2 9 Nd i Nc4 30 Qc3 R a b 8 3I Rei Re4 32 Rc2 Q x d 4 33 Q x d 4 Rxd4 34 R e i a 5 ! F i n e play. 3 4 . . .R b 5 35 Re8+ Kh 7 3 6 Ne3 Rxc5 3 7 Re7 is n o doubt win­ ning but, as usual, it is more effective not to allow such counterplay.

35 Kg2 a4 36 Nc3 a3 37 Na4 g5 38 Re 7 Rb4 39 Nb6 Rb 2 40 Rc3 Rxa2 4 I Nxc4 d x c4 4 2 Rc7 Rdd2 4 3 Rf3 c3 0-I 44 R3xf7 Rxf2+ 45 Rxf2 Rxf2+ and . .. c3-c2 queens. A cl assy game, a n d exactly t h e sort of th ing Black a i m s for with . . . exd6.

Statistics Black has scored about 46%.

Theoretica l ? Not really.

Theoretica l Statu s Extremely soli d . Wh ite definitely needs a more cunning approach .

117

Starting Out: Alekhine's Defence

Game 27 0 Adams • Baburin Kilkenny 1997

1 e4 Nf6 2 e 5 N d 5 3 d4 d 6 4 c4 N b 6 5 e x d 6 e x d 6 6 N c 3 Be7 7 h 3 (Diagram 5)

Diagram 5

Diagrams

White protects the g4-square

De cision time for the knight

Just as i n the .. . cxd6 line, Wh ite wants t o prevent Black's develop­ ment of the bish op on g4.

7

...

0-0 8 N f3 Bf5

8 . . . c5 is an alternative.

9 B e 2 Bf6 1 0 0-0 Re8 10 . . . Nc6 is normal, with play similar to the prev ious ga me. Another plan for Wh ite after 1 0. . . N c6- quite an annoying one, indeed- is 1 1

d5 N e5 1 2 N d4 followed by b2-b3 and Be3 with a space advantage that is v ery diffi cult to sh ake. As Baburin -wh o has great experience

with this system- explains in h is notes, he wanted to try out a new plan. Black can also proceed with 10 c6/a5/N a6 or with 10 . . . N 8d7 fol­ lowed by . . . c7 -c6 and . . Re8, as in Serper-Dreev, Soch i 1986, waiting for a more convenien t moment a t wh ich to play d6-d5. . . .

.

. ..

1 1 Re i Baburin's ideas were to meet 1 1 b3 with 1 l . . . c 5 1 2 Be3 Nc6 1 3 Rc l , or 1 1 Be3 with 1 l . . . d 5 12 c5 Nc4 1 3 Qb3 N xe3 14 fxe 3 Rxe3 15 Qxb7 N d 7. N eith er h as been tested .

ll...Nc6 1 2 a 3 With t h e two rook moves t o the e - file included Wh ite needs t o prevent . . . Nb4 l i k e th is, wh ich must surely help Black, although a2-a3 may h ave a positive side by supporting a later b 2 - b 4 .

1 2... d 5 1 3 c 5 (Diagram 6) 1 3 . . . Nc8 Later Baburin preferred 1 3 ... N c4 14 b3 N4a5 15 Be3 b6.

118

Exchange Variation with 5 . . exd 6 .

1 4 Bf4 a6 14 . . . b6 meets with 15 Bb5 but wh y Baburin rejected 14 . . . N8e7 I'm not quite sure. The obv ious try is 1 5 Nb5 Rc8 16 g4 Be4 1 7 g5 N g6 18 Bxc7 Rxc7 19 Nxc7 Qxc7 20 gxf6 Qf4 b u t Black's compensation h ere seems to be at least a s good a s i n the note to Wh ite's sixteenth move, below.

15 Qd2 N8e7 16 Bfl Wh ite could h ave won a piece here with 1 6 g4 Be4 17 g5 N g6 18 gxf6 Qxf6 19 Be5 Q f5 which Baburin says is unclear, al though I don't think computers would agree. I suspect th ere was an element of bluff here - t h i s was a game played at a relatively quick time l i m i t . Appar­ ently Adams rejected th is a l mo st immediately, displaying a pragma­ tism wh ich p roved its value in th e game. ,

1 6 . . . h6 17 b 4 B e 6 18 N e2 (Diagram 7 )

Diagram 7 The knight relocates to the kingside

Diagram Crafty

B

1 8 ... Ng6 1 8 . .. Nf5 looks th ematic but blocks the e6·bishop and ach ieves li ttle.

19 Be3 Bf5 Now that the k n ight h as left c3 Black returns to the idea of .. .Be4. Baburin m e ntions the quiet 19 ... Qd7 as the alternative.

20 Rad l Q d 7 2 1 Ng3 Be4 2 2 Nh2! Nh4 23 Ng4 Bd 8 24 Ne5?! An oversight; 24 Bf4 w a s better.

24 ... N x e 5 25 d xe5 Bf3! (Diagram 8) Without th is Black would b e i n trouble, b u t now th e game i s unclear .

2 6 R e i R x e 5 2 7 Bxh 6 Rxe l 2 8 Rxe l d 4 2 8 . . .gxh6 29 Qxh 6 B e 4 30 Nxe4 dxe4 3 1 Rxe4 Ng6 3 2 Bc4 was h ope­ less. After the text Black is still i n the ga me, but the clock was becom­ ing the decisive factor.

1 19

Starting Out: Alekhine's Defence

2 9 B g5 Bc6 3 0 Bxd8 Rxd8 31 Qg5 Ng6? Baburin says that 3 l . . .d 3 was better and implies that Black then stands well. He gives the splendid line 32 b5 (to prepare 3 2 Qxh4 which, if played at once, loses to 3 2 ... d 2 3 3 Rd 1 Ba4) 32 ... d2 33 bxc6 dxe1N! 34 Qxh4 bxc6 and Black wins. H e doesn't say, however, how he wou l d have met 32 Bxd3. 32 . .. Qxd3 see m s to be forced (32 .. . Nf3+ 33 gxf3 Qxd3 34 N f5 g6 35 Re6 ! , and 32 ... N xg2 33 N f5 g6 34 Re7 Qxd3 35 N h6+ and m ates, are both out of the question) and after 3 3 Qxh4 Black faces an uphill struggle so far a s I can see.

3 2 h 4? This serves only to lose a tempo and 32 Bc4 was stron ger. White's at­ tack is very dangerous then, for ex a m p le 32 . . . Bb5 3 3 Bb3 Kf8 34 Nf5 d 3 35 Re7 Qxe7 36 N xe7 N x e7 37 Qf4 N d5 38 Qe 5 N f6 (38 . . . d2 39 Qd4 rounds u p the d - pawn) 39 Qxc7 etc.

32 ... d3 3 3 Rd 1 d2 3 4 Bc4 White is worse and possibly lost, but with this move Adams i ns t a n tly rejected a draw offer.

34 ... Q d 4 Baburin correctly rejects 3 4 ... Ba4 35 Qxg6 Kf8 ( 3 5 . . . Bxd1 3 6 N f5) 36 Q h 7 with a decisive attack, but this move cost him half of the two m i n u tes h e h a d l e ft to fin i sh the whole game.

35 h5 N e 5?? 3 5... Qh4? 36 Bxf7+ was another way to lose, but after 35 ... Nh4 Black is doing well, e .g. 36 N fl Bxg2 37 Nxd2 Bf3 (I suspect 3 7 . . . Bc6 is better) 38 Re 1 Qg4+ (38 . . . Qxd2? 39 Bxf7+ wins) 39 Qxg4 Bxg4.

36 Nf5 Qg4 3 7 Nh6+ 1-0 A great scrap which shows Adams' tremendous practical strength.

Stati sti cs/Theoretica I?/Theoreti ca I Status Much the s a m e as Game 26. The many move orders m a k e t h e statis­ tics an unreliable guide in this line, and in any case this system throws the m a i n weight of t h e struggle i n t o understanding of the quiet middlegam e. Game 28 0 Emms • Davies Southend 2002

1 e4 Nf6 2 e 5 Nd5 3 d4 d6 4 c4 N b 6 5 exd6 exd6 6 Nc3 Be7 Assu m i n g that White wants to a dopt the system shown in this game, there are some move order issues already. For exa mple, a t this point Black may choose 6 . . . Nc6 to prevent 7 Bd3. White can try 7 d 5 Ne5 but 8 f4 N ed7 9 Qd4 now seems to be weak- 9 . . . Be7 1 0 Qxg7 B f6 1 1 Qg4 Bxc3+ 1 2 bxc3 Qf6 1 3 Bd2 N c5 gives Black excellent compensa­ tion for his pawn. The more sensible 7 N f3 Bg4 8 Be2 Be7 9 d 5 Bxf3 10 Bxf3 N e5 1 1 Be2 0 - 0 produces a position w h i c h Baburin h a s been

120

Exc ha nge Variat i o n w it h 5 . . exd 6 .

happy to defend, although the textbooks consider i t a little better for White. Black's plan is ... Bf6 , . . . a 7-a5, . .. Ne5-d7-c5, ... ReS and s o forth, although it does seem that White will always be a little better. While Black has little positive to keep him warm, the advantage o f the two bishops is m itigated by the strength of Black's king's bishop, and if this is exchanged for its counterpart the remaining white b isho p is not a great piece. Alternatively Black could try 7 . . . Be7 , when S d5 Ne5 9 N xe5 dxe5 10 Be2 0-0 1 1 0-0 f5 produces a fighting game, al­ though Richard Palliser assures me he h a d the better of the draw in Palliser-Miles, Scarborough 2 00 1 . Returning to move 7 after 6 N c3 Nc6 , White could also think about 7 h 3, returning t o the system in Game 2 7 . Black's ea rly ... Nc6 is not the most flexible way to tackle that, and in particular the d4-d5/Nd4 plan mentioned in the notes to Black's tenth in Game 27 gains in strength. White might try 6 Bd3 when Black can then try to cross Wh ite's plans with 6 ... Nc6 7 Ne2 (7 Nf3 Bg4 i s annoying) 7 . . . N b 4 which wins the two bishops, or perhaps 7 ... g6 now that White cannot arrange Bg5.

7 Bd3 (Diagram 9)

Diagram

Diagram

9

Preparing Nge2

10

Has White made progress?

A trendy system, reco mmended by both Emms a n d Davies in their books, which gives the present encounter a certain piquancy. White's plan with Bd3 a n d Nge2 is again directed against Black's queen's bishop, which is to be prevented from going to f5 and from pinning the knight (thanks t o f2-f3) should it venture to g4.

7 0-0 S Nge2 Nc6 9 0-0 ReS ...

The traditional methods are to seek the exchange of the d3- bishop either by 9 . . . Bf6 1 0 Be3 Nb4 1 1 b3 Nxd3 1 2 Qxd3 Bg4 1 3 f3 Bh5 14 N g3 Bg6 or 9 . Bg4 1 0 f3 Bh5 1 1 Nf4 Bg6 1 2 Bxg6 hxg6 1 3 d5. In either case Black has a defensive task ahead of him, although there is no reason for him to lose. N ote how White makes n o great effort in either line to preserve his king s bishop - in the structure after d4-d5 this is ..

'

121

Starting Out: Alekhine's Defence

not his most effective minor piece . 10 b3 Davies had recommended 9 ...R e8 in his boo k.

so

we can imagine tha t

E m ms was ready for it. The text is better than the previously played 10 a3 - White has no ne ed to preserve the bishop. Mter that move Black can pla y 10 ...Bf6 11 Be3 Bg4 (or at once ...g7-g6. perhaps fol­ lowed by ...N e 7-f5) 12 Qc2 g6 13 h3 (13 f3?? Rxe3 is one of the points of9 ...R e8) 13 ...Bxe2 14 Nxe2 Bg5 15 d5 (15 Qd2!?) NbS! (to ensure the exchange of dark -squared bishops) when Black drew e asily in Shaw­ Davies. 4NCL 2002. 10...Bf8 Bl ack could follow a similar plan to Shaw-Davies. altho ugh 10 b3 may enable White to repserve his dark-squared bishop by developing it on b2. but Davies prefers a different approach. 11 h3 Nb4 12 Be3 c6 13 Bbl A ga in neither side is terribly bo thered about the possibility of ...Nxd3. 13... d5 1 4 a3 Na6 15 c5 Nd7 16 b4 Nc7 (Diagram 10) It lo o ks as tho ugh White has achieved his usual space advanta ge . but in fact Black has organized himself rather well. 17 Bd3 b6 This move is needed to prevent b4-b5 by exerting pressure

on

c5.

18 Qb3 Ne6 19 Rfdl a6 20 c xb6 Nxb6 21 Na4 Nxa4 22 Qxa4 Bd7 23 Nf4 g6! Another instructive move, designed. as Davies says. to prevent the exchange of his 'bad• que en's bishop by Nxe 6 Rxe 6. Bf 5. As often hap­ pens in the QG D Exchange variation. Black needs his bad bishop to def end his go od pawns. 24 Nxe6 Rxe6 25 Qc2 Bd6 '12-% An unexciting draw. but a significant encounter. Black's 17th and 24th moves are instructive .

Game29 0 E.Berg • J.Ericsscn Skara 2002 1 e4 N£6 2 e5 Nd5 3 d4 d6 4 c 4 Nb6 5 exd6 exd6 6 Nc3 Be7 7 Qf3 (Diagram 11) A very aggressive system. although probably no t that go od. It can also be introduced with 6 Qf3. although 6...N c6 7 Be3 Qh4 is then disrup­ tive. so usually White waits until move 7. Once again White is con­ centrating his fire on Black•s poor que en's bishop. which is to be barred from g4 and f5. 7 ...Nc6 7...0-0 8 Be3 c6 followed by ...d6-d5 and (after c4-c5) ...b7-b6 is another solution.

122

Exchange Variation with 5 . . . exd6

Diagram

11

White c l e ars the way for 0-0-0

Diagram

12

Uncompromising play from Black

8 Be3 0-0 9 0-0-0 f5 (Diagram 1 2) Black can also play 9 . . . Be6 at once with similar play - and frequently does - but I like th e space-grabbing text. 9 . . . Bg5 1 0 Nh3 Bxe3+ 1 1 fxe3 Qh 4 12 g3 Qg4 is also fine.

1 0 g3 Berg had played the alternative 10 c 5 before (this is a com m on the m e i n th is system, driving t h e knight to d7 to block the bishop) but a m onth before th is game h ad run into 10 . . . f4 ! ( 1 0 . . . Nd7 is good also) 1 1 cxb6 fxe 3 1 2 bxc7 Qxc7 1 3 Qd5+ ( 1 3 Qxe3 Nb4 1 4 Kd2 Rxf2 + 1 5 Qxf2 Bg5+ 1 6 Ke 1 Qe7+ 1 7 Be2 N c2+ a n d perpetual is a pretty line, or a good i l lustration of why com puters are ruining the game, de pending on your point of view) 1 3 . .. Kh8 1 4 fxe3 Bf5 15 N f3 N b4, and had been forced to bale out with 1 6 Qc4 Qxc4 17 Bxc4 Rac8 18 Be2 N xa2+ with a fairly equal position ( 1 6 Qb3 a5 is dangerous). The untr ied 10 Nh3 i s also worth th inking about.

10 ... Be6 1 1 b 3 After 1 1 c 5 Black sh ould b e brave and go 1 l . ..Nc4 with good play. l l . . . Qc8 Directed against 12 c5, which can be m e t by ... dxc5 and

. . . Nd7-e5.

1 2 Nge2 a5 (Diagram 1 3) 13 a4 13 d 5 Ne5 14 Qg2 Bf7 15 Nf4 h as been s u ggested a s better for Wh ite, but I doubt it. Wh ite's k i n g looks decidedly shaky to m e, and Wh ite's fianch ettoed queen is h ardly radi ating power.

13 ... Bf6 14 N f4 Bf7 1 5 h4 Nb4 16 B h 3 c5 1 7 N fe 2 e x d 4 (D iagram 1 4 ) 18 Bxf5 This loses, but I don't t h i n k the grandmaster blundered - rather he didn't fancy the headlines that might have been made after 18 Bxd4 Bxd4 19 Nxd4 N xc4 20 bxc4 Q xc4. Wh ite i s already in grave diffi culty.

1 23

Starting Out : A lekhine' s Defe n ce

Diagram

13

Now Black switches flanks

Diagram 1 4 Opening t h e c-file

18 ... Be6 19 Bxe6 Qxe6 20 Rxd4 20 Bxd4 Bxd4 2 1 N xd4 Rxf3 22 N xe6 Rxc3+ and Black wi ns.

20 .. . Bxd4 21 N x d 4 Q g 6 A good move. Black is n o t i n fact winning a s easily a s a l l that now that h is attacking momentum is spent.

22 Qe4 N d 3+ 2 3 Kd 2 Nxf2 24 Qxg6 hxg6 25 Rc1 Rae8 26 N d b 5 N c B 2 7 N c 7 N e 4+ 2 8 Nxe4 R x e 4 2 9 Kd3 R e 7 3 0 N d 5 R d 7 3 1 B f4 N e 7 3 2 N x e 7+ R x e 7 33 Kd4 Rf5 34 Rc3 Re 1? Black has played sensibly and i s winning, but no doubt the earlier complications had led to t i m e trouble, and th ese GMs can be dam ned elusive. 34 . . . Re6 was better planning to bring the king to the centre.

3 5 Rd 3 Re6 36 Kc3 Rff6 3 7 Rd5 Rf5 38 R d 4 Rff6 39 Rd5 Rf5 40 Rd4 Rff6 Yiz-ih A disappointing en d, but since Black wa s playing u p by 220 ELO points against a n expert i n th is line, he couldn't h ave been too sad.

Statistics Wh ite h as scored 53%.

Theoretical? A bit. It's h elpfu l to have an answer in m i n d .

Theoretical Statu s Caused panic for a wh ile b u t I th ink Black has the h a ng o f i t now.

1 24

C ha pter S ix

F o u r P a wn s Atta c k

• •

I n trod u ction I l l ust rative G ames

Start i n g O ut: A lekhine' s Defe n ce

l ntrod uction 1 e4 Nf6 2 e 5 N d 5 3 d4 d 6 4 c4 N b 6 5 f4 (D iagram 1 )

Diagram

Diagram

1

White's broad pawn centre

2

A popular position

The Four Pawns Attack is the Alekhine Defence equivalent of the K ing's G a mbit - to quote John E m m s: i t doesn't matter how long you've been playing the opening, that lunge with the f- pawn still sends a shiver down your spine. White's idea is agreeably simple: he is going t o h o l d on to the cramping pawn on e5 a n d attem pt to batter his way to the fu l l point. Nevertheless , Black has an equal share of the fun and several ways, according to taste, of obtaining a reasonable ga me. ('When I saw 4 c4, I knew what the next move would be. 5 f4 is n o t something top players go i n for' - Alexander Baburin, com ment· ing o n h is game with Ada ms - Ga me 27. He was right - 5 exd6!)

5 ... dxe5 Black has various m o re o r less lunatic alternatives at this point, in­ cluding the absurd-looking 5 . . . g5, which does have its own crazy logic and is not half as simple to refute as it l ooks. This is covered in Game 34, with 5 . g6 a n d Alex Wahl's 5 . . . c5!? in the notes. It is also possible to play 5 . Bf5, which is part of some im portant transpositional tricks which are covered in the notes to Game 3 2 . .

.

. .

6 fxe 5 N c 6 6 . . c 5 is a fasc i n a ti n g alternative with various semi- respectable ideas .

in m i n d - this is covered i n Game 33.

7 Be3 7 Nf3 is m uch less good as Black is able to apply i m mediate pressure with 7 . Bg4 . White may then venture 8 e6? ! but Black can sim ply take th is with the f-p awn a n d develop via g 7 - g6 and so on. This i s w h y Black usually play s 6 . N c 6 before Bf5: he wants to force B e 3 . . .

. . .

. .

126

. . .

Four Pawns Attack

7

...

Bf5 8 N c3

8 N f3 N b 4 forces the undes irable 9 N a3.

8 e 6 9 Nf3 (D iagram 2) . . .

Both sides have developed naturally and this position is the starting point for m uch of the theory. White m ay of course carry on develop­ ing, but n o w that he has e5 and d5 covered h e is also ready to lunge with d4 - d 5 followed by Nd4. Black has several m oves here which are designed to prevent that. I n fact 9 . . . Bb4, 9 . . . Bg4 , Alekhine's original 9 . Nb4 and even 9 . Qd7 (via various obscure tactical am bushes) all do this, albeit with different degrees of theoretical a pproval . These are a l l covered in Gam e 3 1 . The m ain line, however, has long been . . . ..

.

.

9 Be7 . . .

... and t h a t is the m ove w e focus on in G a me 30. This does not prevent 10 d5 - on the contrary it challe nges White to p lay it i f he wants to s e ek a n advantage. I t m ust be s a id a t the outset that although Black has decent chances theoretically this is not a com fortable solution for the casual player. The positions which arise are com plex a n d well­ analysed, a t least i n areas ( people were swapping innovations on move 2 3 i n early 1 9 70's Cambridge circles). Black has to tread a nar­ row path, and for a rare line like the Four Pawns Attack a reply which leads to less explore d p l ay might be preferab le. Moreover there are a couple of virtually forced draws lurking in the notes to Game 30.

ustrative Games Game 30 0 Hoskyn • Dive New Zealand Correspondence Championship 1 997 1 e4 N f6 2 e5 Nd5 3 d4 d6 4 c4 Nb6 5 f4 dxe5 6 fxe5 N c 6 7 B e 3 Bf5 8 Nc3 e6 9 Nf3 Be7 (Diagram 3)

Diagram

3

Black provokes d4-d5

Diagram

4

Calmly does it. . .

1 27

Starting O ut : A le khine ' s Defence

10 d 5 White c a n play more tam ely at this poi nt with 1 0 B e 2 . The usual se· quence after this, which has been played in many games, is 10 . . . 0-0 1 1 0 - 0 f6 1 2 exf6 (the decentralizing 1 2 N h 4 i s not worth i t but Black has to find 12 ... fxe 5 13 Nxf5 exf5 14 d 5 Nd4! , when 1 5 Bxd4 exd4 16 Qxd4 Nd7 gives him da ngerous play on the dark squares ; 1 2 a 3 fxe5 13 d 5 c 4 i s also good for Black) 1 2 . . . Bxf6 1 3 Qd2 Qe7 (the best place for the queen; Black would not wish for a n exchange of the dar k - s q u a r e d bis hops , and this move enables h i m to meet 1 4 Bg5? with 1 4 . . . Bxg5 1 5 Nxg5 N xd4) 1 4 Rad 1 Rad8. Korchnoi declared i n Chess Is My Life that White stood better here, b u t in fact practice h a s shown that Black i s fine. His next few moves will typically be ... h7-h6 (ruling out Bg5 once and for all) ... Kh8 (the king i s safer here in the event of d4d 5) ... Bh 7 (also directed a ga i n st d 4 - d 5 followed by N d4) a n d ... Rfe8, i n some order o r other. Black then has e veryt h in g organized, his e-pawn i s n o weaker than White's d-pawn, a n d h e can cons ider s uch ideas a s . . . Rd7/Red8, . . . Qf7 (if a white rook comes to t h e e-fil e) a n d perhaps . . . N c8-d6, i m proving his worst pi ece. The m ain pitfall to beware is playing . . . e6-e5 premature ly - for some reason in m a ny gam es Black has m iscalculated and made this central break i n positions w h e re it loses .

1 0 ... exd5 The older 10 ... Nb4 leads to spectacular play a fter either 1 1 Rc 1 or 1 1 Nd4. The consensus i s that White i s better, a n d although the l ast word probably has not been s a i d I have not pres ented the material i n v i ew o f t h e s oundness o f t h e text.

11 cxd 5 N b 4 The only s ensible square.

12 N d 4 1 2 Bxb6 (or a m ove earlier) h a d a period o f popularity but it i s not a terribly good idea. The point is to rem ove the threat to the d -pawn so that after 12 . . . axb6 13 N d4 if B l ack plays as in the m ai n ga m e with 13 . . . Bd7 h e is hit with the strong 14 a 3 ! , driving the knight away a n d consolidating. Ins tead o f fa lling for this Bl ack can either venture Ber­ n a r d Cafferty's cute and untried 13 . . . Be4, with the point that 14 N xe4 Qxd5 forks the two knights ( . . . Qxd4, Qxd4 Nc2+), or play the well · tested 1 3 . . . B g 6 . T h e m ain difference from t h e text i s that 1 4 Bb5+ c6 15 dxc6 0-0 16 cxb7 Rb8 (or 1 6 . . . Ra5) is now good for Bl a c k due to the exchange o f his poor k night on b6 for the valuable dark- squared bis hop. After 14 d6 all of 14 . . . Bxd6, 14 . . . 0 - 0 and 1 4 . . . Bh4+ lead to com plications i n wh ich White is supposed to be struggling, and cer­ tainly White has abandoned this line i n recent years.

12 . . . B d 7 For a move which virtually loses , 1 2 . . . Bg6? h a s been surprisin gly popular. 13 B b 5+ then forces either an inadequate pawn sacrifice with 13 . . . c6 14 dxc6 0 - 0 15 cxb7 or the miserable 1 3 . . . Kf8 1 4 0 - 0 Kg8,

1 28

F o u r P a w n s Atta c k

after w h i c h Black w i l l h a v e to play the rest of the g a m e without his k i n g's rook.

13 e 6 White was at the crossroads. The d - p aw n i s en prise, s o there are two reasonable a lternatives, nam ely 13 Qb3 and 13 Qf3. The latter l o s t a grea t deal of popul arity a fter White was r ent limb from limb in the s plendid game Pegoraro-Henderson, Ischia 1 99 6 , which proceeded 13 . . . c5 14 dxc6 (for ced in view of the threat of . . . Nc2+ if the d 4 - knight moves) 1 4 ... bxc6 ( 1 4 . . . Nxc6 looks good until you s ee 1 5 e6! fxe6 16 Bd3 catching Black's king horr ibly in the centre) 1 5 e6 ( 1 5 a3 c5! 16 axb4 cxd4 17 Bxd4 0 0 ! and not 1 7 ... Bxb4 1 8 e6! ; 1 5 . . . N4d5 16 Nxd5 Nxd5 1 7 Bc4 i s m ore dangerous with the d4-knight sti l l a live) -

-

T I P : In this variation B l ack freq uently u s e s the counterattack on the d4-knight to avoid h a v i n g h i s own k n i g ht d riven away from b4.

15 . . .fxe6 16 0-0-0 (now 16 a3 N 4d5 is the better move s i nce a fter 16 . . . c5? 1 7 axb4 cxd4 18 Bxd4 Black cannot castle) 1 6 . . . N6d5 1 7 a3 Nxc3 1 8 Nxe6 (White takes the opportunity to play this w h i le he c a n - 1 8 bxc3 Nd5) 18 . . . Nca2+ (White's very natural attacking s equence had led to victory and a brilliancy prize i n an earlier gam e after 1 8 . . . Qa5 19 Nxg7+ Kd8 2 0 Rxd7+; Caffer ty h a d recomm ended this queen sacrifice in BCM, but evidently Mr Pegoraro was not a s ub ­ scriber) 19 Kb 1 Bxe6 20 Rxd8+ R x d 8 21 axb4 Nxb4 2 2 Be2 Rf8 2 3 Qh5+ g6 2 4 Q e 5 Bf5+ 2 5 Ka 1 Rd5 26 Q b 8 + Kf7 0 1 . White i s about t o be s laughter ed by 2 7 Q x a 7 N c 2 + 2 8 Ka2 N x e 3 2 9 Qxe3 R a 5 + 30 Kb3 Rb8+ etc. You can see why that put Whites off. -

1 3 Qb3 can be met in s im ilar fash ion with 1 3 . . . c5 1 4 dxc6 bxc6 1 5 0-00 Qc7, although 16 e 6 fxe6 17 Nxe6 Qe5 1 8 Rxd7 h a s been touted as a possibility, which rem ains unclear.

13 ... fxe 6 14 dxe6 Bc6 1 5 Qg4 15 Nxc6 Qxd 1 + 16 Rxd 1 N c2+ 1 7 Kd2 N xe3 18 Kxe3 bxc6 is equal and drawish - the e-pawn is a liability a s m uch as an asset and in view of the opposite colour ed bishops can never be ferried across e7. 1 5 Qh5+ g6 16 Qh6 Bh4+ 1 7 g3 Bf6 has also been tried, but now that Black has r uled out Nf5 a n d es tablis hed a decent hom e on e 7 for h is queen he s hould be comfortable.

15 ... Bh4+ Pretty m uch the only m ove: Black needs both to deal with the attack on his g- pawn and to prevent a devastati ng 0-0-0.

16 g3 Bxh 1 17 0-0-0 17 Bb 5+ c6 18 0 -0- 0 0-0 19 gxh4 is m e t by the cunning 1 9 . . . h 5 ! - a com mon them e in this l i ne. Black cannot take on b5 as long as White can reply 20 Nf5 (threatening mate on g7) 20 . . . Qf6 21 Bd4, but i f he can deflect the q u e e n first from g7, then he can. It a l so tur n s out that 2 0 Qg3 cxb5 21 Nf5 can be m et by 2 1 . . . Rx f5 22 Rxd8 Rxd8 with too m uch m a ter ial for the queen, so Black survives after (for example) 20

1 29

Sta rti n g O ut : A l e k h i n e 's D e fe n ce

Qg3 cxb5 2 1 Bg5 Qb8, forcing t h e queens off.

1 7 . 0-0! (Diagram 4) ..

The greedy 1 7 . . . Bf6 runs into the crushing 18 Nf5.

18 gxh4 Q f6 1 9 Bb5 White has various options including a practically forced draw with 1 9 B g 5 Qxfl 20 Rxfl Rxfl+ 2 1 N d l N d 3+ 2 2 Kc2 N e l + , a n d 1 9 Be2, when 19 ... Bd5 i s the recom mended move for a reason we shall s ee in the note to White's 2 3rd m ove. The idea o f the text is to control the promotion s quare of the e-pawn.

1 9 . . . c5 2 0 Bg5 White continues with his plans for the e-pawn, but there was s ome­ thing to b e s aid for baling out with 20 Rxh 1 cxd4 2 1 Bxd4 Qf4+ 2 2 Qxf4 Rxf4 2 3 Bxb6 axb6 2 4 e 7 Nc6 2 5 Nd5 Re4 2 6 Rg1 when White has enough play to win his excha nge back, but little more. 20 ... N6d5 2 1 Bg5 Qxd4 22 Qxd4 cxd4 23 Nxd5 Rac8+ 24 Kb 1 Nxd5 25 e7 Rfe8 with a draw may be even more accurate.

20 ... Q e 5 2 1 e7 c x d 4 22 e xf8Q+ Rxf8 23 Rxh 1 Black had a little trick here - the natural 2 3 Qxd4 is met by 2 3 ... Nxa2+ 2 4 Nxa2 Qxb5 2 5 Rxh 1 ? Qc6+. Had White played 1 9 Be2 and play proceeded similarly then 23 Qxd4 Qxd4 24 Rxd4 would have led to a sem i-ending which is really qu ite unpleasant for Black to de­ fen d .

2 3 ... h 6 ! (Diagram 5)

Diagram 5 P recise

Diagram

6

Black emerges unscathed

23 . . . dxc3 2 4 Qxb4 cxb2+ 25 K b 1 Nd5 leads to an attack which is good for perpetual check for one s ide or the other, b u t Black is playing to win. Mter the text White's g5-bishop has no good square.

24 B d 2 2 4 Bxh6 dxc3 2 5 Qxb4 cxb2+ 26 Kb 1 Qf5+ 27 Kxb2 Qf6+ and Qxh6 is

1 30

F o ur P a w n s Atta c k

the point. Black stands decidedly better then, but this w a s preferable to the game.

2 4 ... dxc3 25 Bc4+ Forced in view of the threat of 25 . . . Nxa2+.

25 ... Nxc4 2 6 Qxc4+ Kh8 27 bxc3 Worse for White i s 27 Bxc3 Nxa2+ 2 8 Qxa2? Qe3+, allowing Black to harpoon the rook o n hl next move (29 Bd2 ReS+).

27 ... Nc6 28 Rfl Rd8 ( D iagram 6) Naturally Black keeps the rooks on in view of White's unsafe king. The position has stabilized and Black has a definite advantage with his better pawns and safer king.

29 Rf2 a6 There is no hurry. White cannot plug the gaps.

3 0 Re2 Qd6 3 1 Qe6 Qc5 32 Qf7 Ne5 33 Qxb7 Nc4 3 4 Qb3 Nxd2 35 Rxd2 Qe3 36 Qc2 h5! 0 - 1 A sweet finish. None of White's pieces c a n move, a n d when h i s p a w n moves r u n out it will be t h e e n d .

Statistics Reassuring. Black has scored 46% after 1 2 . . . Bd7, and 49% after the quieter 10 Be2.

Theo retical? I am afraid so. If Black doesn't know what he i s doing u p to move 1 8 i n the ga m e there are several horrible pit fa ll s a n d even after that a t least h a vin g seen the tactics b e fore i s handy when i t comes t o trying to work them out. ,

Theoretical Status Excellent. No doubt there is m o r e to be discovered, but this variation has been heavily investigated for thirty years and if there i s one re ply to the Four Pawns Attack which i s not going to be refuted, this is it. Game 31 0 R.Byrn e • Albu rt Berkeley 1984

1 e4 Nf6 2 e5 Nd5 3 d4 d 6 4 c4 Nb6 5 f4 dxe5 6 fxe5 Nc6 7 Be3 Bf5 8 Nc3 e6 9 Nf3 Bg4 This is certainly Black's first reserve to 9 . . . Be7 b u t , as I said before, Black has a number of other moves here. Perhaps the longest estab­ lished is 9 Qd7. Black prepares either .. 0-0-0 o r ... Rd8 with pressure o n the d-pawn, and also readies h imself for 10 d 5 by defe n ding the bishop on f5 . White can still play 10 d5. Black continues 10 . . . exd5 1 1 cxd5 Nb4 1 2 N d 4 a n d now he has two po ss ibil i t i e s 1 2 . . . N6xd5 1 3 . . .

.

.

131

Starti n g O ut: Alekhi ne's Defe nce N x d 5 ( 1 3 Nxf5? Qxf5 1 4 N x d 5

0-0-0 or the i m m e d i at e 1 3 . . . 0-0-0 are 0-0-0 1 5 Q d 3

very d a n gerous for White, if not lost) 13 . . . Nxd5 14 Nxf5

g6 produces fearsome a n d untried com p!ice.t; o::m"', as does the once ­

c ondem n e d 12 . . . Bg6! ? . In the marvellous recent biography B e n k o points o u t t h a t t h e 'refutation' 1 3 Bb5 c 6 1 4 d x c 6 b x c 6 1 5 N x c 6 ( ! ) N x c 6 1 6 Qf3 R e S 1 7 Rc l a c t u a l l y l o s e s t o 1 7 . . . N c 4 ! 1 8 Bf4 B d3 ! ! etc. W hite's ( m u c h c o m m oner) alternative is 10 Be2. 1 0 . . . 0-0-0 here is presently not looking good (for those with databases, the ga mes M i ­ nasian - Donchenko, 1 988 a n d Kotronias - Short, which for re a so n s

o n ly

Gib ral ta r 2003,

in

he coul d e x p la i n Short re pe a te d a l i n e l o n g

k nown t o b e b a d for Black u p t o t h e hideous 1 6 . . . Na8, will s h o w why), so Black is w iser to try

10 . . . Rd8 1 1 0-0 (if White prepares to d e fe n d

t h e d-pawn with 1 1 Q d 2 then l l . . .Na5 is a n noying , hitting the c-pawn a n d prepa ring moves like

. . . Bb4 , . . . Bg4

and perhaps . . . c7-c5 or . . .f7-f6,

putting great pressure on W hite's centre) . After 1 1

0-0 Bg4 the W h ite

d- pawn is en prise. W hite h a s the choice between the sensible 1 2 c5 N d 5 13 N x d 5 Q x d 5 1 4 Ng5 Bxe2 15 Qxe2 Rd 7 1 6 Qd2 Nd8, when

Black i s cra m p e d but solid and can a i m for

...

f7-f6,

and the wild 1 2

Ng5 Bxe2 1 3 Qxe2 (or 1 3 Nxe2) 1 3 . . .Nxd4 1 4 Bxd4 Qxd4+ 1 5 Kh 1 ,

which produces another virtually u n ch arted thicket o f com plications . Another possibility i s t h e developing 9 ... B b 4 , which prevents 10 d 5 radica lly a n d a i m s for cou nterplay v(a eit h e r . . . f7-f6 or . . . Bxc3 a n d . . . Na5. W hite's most sensible reply is 1 0 B e 2

0-0 1 1 Rc 1 , meeting

1 l . . . N a 5 with 1 2 Bg5 Qd7 1 3 c5 Na4 1 4 Q d 2 and so on. White can c a s ­ tle o n the kingside a n d h a s d a n gerous attacking chances there with Black's minor pieces stuck on the queenside . Bl a c k has yet to demon­ strate a decent cou nter to this plan. Finally, 9 . . . Nb4 1 0 Rc1 c 5 w a s Alekhine's i d e a to demolish White's centre, but after t h e simple 1 1 Be2 Black has difficulties, for example 1 l . . .cxd4 1 2 Nxd4 Bg6 1 3 c5! and Qa4+ ( 1 3 . . . Bxc5 1 4 Bb5+ Nd7 1 5 Nxe6 i s dire) , o r 1 l . . . Be 7 1 2

0-0 0-0 1 3 dxc5 N d 7 1 4 b 4 .

The t e x t h a s a d u a l p u rpose, preventing 1 0 d 5 i n v i e w of 1 0 . . . Bxf3 a n d . . . N x e 5 a n d preparing a n a s s a u l t o n the d - pawn with . . . Bxf3.

10 Be2 The main reply. 10 Qd2 i s also possible b u t Black obtains go o d play with 10 . . . Bb4 1 1 a 3 B e 7 ! (the idea of this is to weaken b3 and make . . . Na5 more effective). This h a s proved very effective fo r Black . Hueb­ ne r- H o r t , 1 987 and Fedorov-Baburin, 2000 are the mai n games that show why - i n a nutshell, Black will play . . . Qd 7 and . . . 0-0-0. If White a l s o castles queenside, then h e has troubles with the laser bishop on f5 a n d t h e constant possibility o f . . . N a 5 -b3 m a te , a n d if, on the other hand , he goes for i t with b 2 - b 4 t h e n Black c a n play

. . . Bxf3

a n d again W hite's k i n g finds no rest.

10 . . . Bxf3 The only popular move, preparing the following idea.

1 1 gxf3

1 32

and

...

f7-f6

F o u r P a w n s A tta c k

1 1 Bxf3 N xc 4 .

1 1 . . .Q h 4+ 1 2 B f2 Q £4 ( D i agram 7 )

Diagram

Diagram 7

8

An impressive retreat

A classic 'Four Pawns Attack' position

A m odern -look ing m anouevre first essayed i n Lasker-Tarrasch , 1 92 3 (although B l a c k h a d a n extr a tem po there).

13 c5 13

Qc 1 Qxc l + 1 4 Rxc 1 is l ess pres s i ng, s i nce Black has the t i m e t o 0-0-0 a n d recapture on d 5 with t h e rook after 1 5 c 5 N d 5 .

play 1 4

...

1 3 ... N d 7 1 3 . . .N d 5 i s

also

possible . T h e dernier cri o f theory i n this l i n e is 1 3 c5

Nd5 14 Nxd5 e x d 5 1 5 Qd2 Qxd2+ 16 K x d 2 g6 1 7 f 4 (it is not neces s ar ­ i l y corr ect t o p l a y t h i s a t once - E CO likes 1 7 Be3) 1 7 . . . Bh6 18 Be3 Ne7 19 Bd3 Nf5 20 Bxf5 gxf5 2 1 Rh g 1 Kd7 22 Rg3 Rag8 23 b4 Rxg3,

Ata l i k - d e Firmian, 2002 (im proving on an ear l i er gam e in 2001 be ·

tween Tim m a n a n d D e Firmian where 23 . . . Rg6 was pla ye d a n d Black was s quashed). Now B l a ck c a n constr uct a n i m penetr a ble fortress with . . . h7-h5 and s o on. Both s ides ' play here was very the m atic ­ Black's m anoeuvre of the kni ght to f5 especia lly - b u t nonetheless I fe el this is a s li ghtl y d epressing, i f solid, w a y for Black to play. Once he has to capture on d 5 with a pawn he loses a lot of his pos s ibilities; he c a n hardly expect to win these pos itions and Whi te, no doubt, h a s other m ove o rders which will require equally accurate defence.

14 Qc 1 14 B b 5 is actually a bit a w k w a rd, thr eatening to double the c- pa wns

before Black can get i n . 0-0-0 a n d . . . Ndb8 as i n the m ai n g a m e , for exam ple 1 4 Be 7 1 5 0 - 0 B h 4 1 6 Bxc6 bxc6 1 7 Qa4 0-0 18 R a d 1 Rab8 . .

. . .

19

Qxc6. Black hasn't dem onstr ated a good ant idote

t h e mom ent. A l b u rt's proposal o f 1 4

...

to this plan a t

f6 has not been tried but, l i k e

Davies , I d o n ' t fancy i t m u ch. I f I were B l a c k i n t h i s pos ition I t h i n k I would be i nc l i n e d to try 1 4 . . . N e 7 . This takes the knight towards

1 33

Starti n g O u t : A l e k h i ne's Defe n ce

where it wants to go; Black intends ... 0-0- 0, so to make sense of his 14 Bb5 White has to go 1 5 Bxd7+ Kxd7 but I'm not sure that he can ex­ ploit the resulting dislocation of Black's king, hampered as he is by the need to keep f3 covered. Meanwhile the exchange of White's king's bishop for a knight is a definite achievement for Black, and if White cannot act quickly Black is ready with . . . Rd8 and . . . Kc8. What White can do i s force the exchange of queens in a way which improves his pawn structure with 1 6 Qd3 Rd8 17 Q e 4 Qxe4 1 8 fxe 4, but I think Black's still fine as he can continue with . . . Kc8 and . . . Nc6 to hold back d4-d5 by the pressure on e5, and then ... b7-b6 or . . . f7.f6 to break up the centre. The text is not presently regarded as theoretically crucial, but I have chosen this game as it illustrates well Black's typical plans in these endings, as well as being a fi ne game by one of the great upholders of Alekhine's Defence. 1 4 ... Qxcl + 15 Rxc 1 0-0-0 1 6 Ne4 N d b 8 ! (Diagram 8)

Excellent. Black's knight had no future on d7. Regrouping this knight to c6 and the present c6- knight either to f5 via e7 or to d5 via either b4 or e7 is a typical procedure for Black in this line, often combined with . . . g 7-g6 and . . . Bh6. The thematic alternative i s an early break with .. .f7-f6 and . . . g7-g5 but here White's knight manoeuvre has discouraged that idea for the moment. 1 7 N g 5 Rd7 1 8 Bb5 h 6 1 9 Ne4 g 6 20 0-0 Bg7

In spite of his lack of space and the two bishops Black al ready stands well. White's weakened pawn structure affords both knights superb outposts on d5 and f5, while the d-pawn is constantly weak and the e3-bishop has few prospects. 21 f4 h5

Now that the f- pawn is on f4 Black wants h6 for his bishop. 2 2 Rc4 Bh6 2 3 Be 3 a 6 24 Ba4 Rdd8 (Diagram 9)

Diagram

9

Another constructive retreat

134

Diagram

10

Wrong plan?

F o u r P a w n s A tta c k

Little by little Black un t a n gl es h i m self. H e needs t o u n pin t h e c6knight i n order t o move i t and free c6 for his other k n i g h t. M e a n w h i le, although his position may look imposing, White can only wait.

25 b 4 N a 7 2 6 b 5 requires prevention.

26 Bb3 Nbc6 27 a4 Kb8 28 Kg2 Ne7 T h i s i s typically thematic i n t h i s p a w n structure.

29 Kf3 Nd5 30 Bd2 b 5 ? ! (Diagram 1 0 ) Alburt s e e s a tactical opportunity b u t I wonder i f h e should have con­ tinued m anoeuvring instead .

31 cxb6 W hite cannot allow the blockade of the queens ide. Black could send h i s other knight t o f5 a n d consider how to i m prove further after that.

3 1 . .. Nxb6 32 R c 2 Rxd4 33 Nc5 W h i t e w i n s his p a w n b a c k , a n d Alburt m a y have u n derestim a te d the p l a y W h i t e now gets - see move 35.

33 . .. Ka8 3 4 Ke3 O f c o u rse 34 Nxa6? loses to 34 ... Rd3+, winni ng the bishop o n b3.

34 ... Rhd8 35 a 5? Computers say t h a t 3 5 Bc3 was good for W h i t e - he d r i ves the rooks off the d - fi l e b efore going Nxa6, for e x a m pl e 35

. . .

Rdl 36 Rxd l Rxd l 37

N x a 6 a n d I haven't been able to discover a n y reason why this is n't true.

35 ... N b 5 ! N o w Black is winning.

3 6 a xb6 cxb 6 T h e threat o f . . . R d 3 + recovers t h e piece.

37 Nxe6 37 Bxe6 bxc5 38 Bxf7 g 5 is no better.

37 ... fxe 6 38 Bxe6 g5! 39 B e l T h e threat w a s

3 9 gxf4+ 4 0 K e 2 f3+ winning a piece, a n d W h ite h a d ...

no good defence.

39 ... gxf4+ 40 K£3 Rd3+ 4 1 Kg2 £3+ Alburt finishes ruthless ly. The pawn cannot be taken because of

42 . . . Rxf3 43 Kxf3 Nd4+ etc. 42 Kg3 Bg7 43 Kh4 Bxe5 4 4 Kxh 5 Ka7 45 Bc4 R3d7 46 Bxb5 Rg7 4 7 Bd 2 R h 8+ 4 8 B h 6 Bf4 4 9 Rc6 Rg5+ 5 0 K h 4 0 - 1 W h i t e d i d n ' t w a i t for 5 0 . . . Rxb5.

Statistics Encoumging indeed. Black has scored 51 % with 9 . . . Bg4, a n d still 49%

1 35

Sta r t i n g O u t : A l e k h i n e 's D efe n c e

even after 10 Be2. 9 . . . Qd7 has also m an a ge d 5 1 % , despite a s l i gh t l y dubious reputation.

Theoretical? Not terribly. With 9 . . . Bg4 Black needs to be comfortable w i t h a space disadvantage and better structure, but there's not that m u c h theory to know. 9 . . . Qd7 is extrem ely complica ted, but theory has only scratched the surface.

Theoretical stat u s 9 . . . Bg4 i s slightly controversial. 1 3 c 5 N d 7 1 4 Bb5 i s a threat, but 1 3 c 5 N d 5 m u s t be a bit better for White i n m y opinion, despite d e Fir· mian's eventual success i n drawing the resulting e n d g a m e . T h e 9 . . . Qd7 system i s o n l y l o o k i n g a good bet w i t h 10 . . . Rd8 a t the m o m e n t . 9 . . . Bb4 a n d 9 . . . Nb4 - particularly the latter - are under a dark cloud a n d look like staying there. One s m a ll point t o consider before m a ki n g 9 . . . B g 4 you r main defence to the Four Pawns Attack : W hite can rule it o u t altogether with 9 Be2 ! Thus Black's most sensible procedure is to play 9 . . . Be7 10 N f3 00 1 1 0-0 f6, tran sposing to the quiet system explained in the note to

White's t e nth m ove in

G a m e 30, which is

fine for Bla ck.

Game 32 0 Majostorovi c • Bryson Correspondence 1 985

1 e4 N f6 2 e5 Nd5 3 d4 d6 4 c4 N b 6 5 f4 Bf5 (D ia gram 1 1 )

Diagram

11

Black refrains from . . . dxe5

Diagram

12

Black changes course

The system Black employs in this ga m e has perhaps the best reputa ­ tion of any against the Four P a w n s Attack. One question is whether i t i s actu ally possible to employ i t a t a l l , given a modicum o f move-order

1 36

F o u r P a w n s Atta c k

c u n n i n g b y White, a n d I h a v e twe a k e d t h e m ove order o f this ga m e a bit to m ake m y p o i n t. The key m ove to bear in m in d is Bd3 for White. If W h i t e c a n get a w a y with this without some inconveni ence resulting (basi c all y , either losing the e - p a w n t o N x e 5 with t h e d - p aw n pinned against the q u e e n o n d 3, o r allowing a n effective ... c 7-c5), then h e also gets control of the i m portant e 4 -squa re. A knight there c a n be devas ­ tating. Black has to arrange things in such a way that this m ove does not work, and this is why his m ost traditional m ove order is 5 . . . d xe 5 6 fxe5 Nc6, so as to attack d4 a n d force 7 B e 3 . N ote that B e 3 does n o t contribute t o m aking Bd3 p l ay a b l e s ince it does not defe n d the e ­ p a w n . I n the traditional o r d e r B l a c k then plays 7 . . . Bf5, threatening 8 . . . Nb4 a n d so forcing 8 N c3. T h e n 8 . . . e 6 9 Bd3 s ti l l loses the e - p aw n , s o W h i t e p lays 9 N f3 a n d we h a v e the m a i n l i n e position of G a m e s 30 and 3 1 . However, 6 . . . Nc6 blocks the c - p a w n , a n d Black would l i k e to b e able to u s e that to attack W h ite's centre, s o in the present s y s t e m he de­ lays . . . N c6 and i n s tead walks a tactical tightrope to stop Bd3. For the m ome n t the text is fi ne, the cunning plan being s een i n the continua­ tion 6 B d 3 B x d3 7 Qxd3 dxe5 8 fxe 5 c 5 9 d 5 ? e 6 10 d6/Nc3 Qh4+ win­ ning the c4-pawn.

6 Nc3 e 6 7 B e 3 T h i s i s the crucial m oment. 7 Bd3 i s s till n o good ( 7 . . . B x d 3 8 Q x d 3 c 5 9 d 5 ? d x e 5 1 0 fxe 5 Q h 4 +) b u t w i t h 7 N f3 W hite c a n prevent Bl ack's idea. The point is that after 7 Nf3 dxe5 8 fxe 5 Bb4 (8 . . . Be7 9 Bd 3 is likewise strong for t h e s a m e reas ons) 9 Bd3 Bxd3 1 0 Qxd3 c5 1 1 0 - 0

cxd4 W h i t e i s i n t i m e for 1 2 N e 4 ! , without needing for the m om ent t o

recapture o n

d 4 (clearly i f W h i t e had was ted t i m e on B e 3 he would

not yet have been able to castle and the c3-knight would s till b e p i n n e d a t this juncture). After 1 2 N e 4 W h i t e has v e r y dangerous a t ­

tacking c h a n c e s indeed with his threat of c 4 - c 5 an d Nd6+, a n d the ever-present danger of Nf6+ if Black castles short (when this position a ri s e s B l a c k is generally m ated b efore m ove 2 5 ) .

Black's i nd e p e n d e n t option after 7 Nf3 (with 7 . . . d x e 5 8 fxe5 N c 6 h e can return t o t h e m a i n l i nes , since White h a s no better t h a n 9 Be3; Black is by now ready to ans wer 9 Bd3 wi th 9 . . . Bg4!) i s 7 ... Na6

(Dia­

gram 1 2). Bla ck's idea with t h i s move i s to challenge the centre w i t h . .. c 7-c5 which , i f played a t once, would fail to 8 dxc5 dxc5 9 Qxd8+ with a dire endi ng for Black. His ideas a r e well s een after 8 Be3 c5, for exa mple 9 d5? N b4 1 0 Rei dxe5 1 1 Nxe5 f6 chasing the k n ight away a n d wi n ­ n i n g the d 5 - pawn, o r 9 Be2 B e 7 10 0-0 0 - 0 1 1 Qd2 cxd4, while 8 Bd3 Bxd3 9 Qxd3 c 5 i s already a t least equal for B l a ck . White h a s two s ensible ways to play. O n e i s 8 Be3 c5 9 dxc5 dxc5 (9 ... Nxc5 is m ore

comba tive a n d v i r tually

untested)

10 a3 (directed against the a 6 -

knight) l O . . . Qxd l + 1 1 Rxd 1 B e 7 1 2 Be 2 w i t h a tedious but defens ible pos ition for Black. The other, however, is the real problem with this line i n m y view, and that i s (7 . . . N a6) 8 exd6! . Both Burgess a n d D a -

1 37

Start i n g O u t : A l e k h i n e ' s Defence

vies suggest 8 . . . Nb4 now - Davies quite enthusiastically - but so far as I can see this does not offer sufficient compensation after 9 dxc7 Qxc7 (9 . Qc8 10 c5 and 1 1 Bb5 is much worse) lO Nb5 Qd8 1 1 Na3. Consequently I think Black is forced to continue 8 ... cxd6 9 Be2 Be7 10 Be3 0-0 1 1 Rc 1 ! (and not 11 0-0 when Black can obtain some freedom with 1 l . . .Rc8 12 b3 Nd5!). This produces a curious Exchange-type setup where on the one hand White has played the rather uncalled-for f2 -f4 and on the other Black has developed his knight to the horrid a6-square. I suspect this favours White as the f4-pawn is not useless in that it restrains . . . e6-e5, whereas the knight on a6 is just terrible and will take a couple of tempi to find somewhere sensible, time that White can use to further his queenside pawn push. . .

Thus we can conclude that Black's move order is only a trick - if White plays 7 Nf3 Black has no good independent choice. I t is quite a useful trick to know for all that, certainly for White, because the sys· tern in this game is worth avoiding, if only to restrict Black's options. 7 . . . dxe5

8 fxe5 Bb4 9 Nf3

9 a3? Bxc3+ 10 bxc3 Qh4+ is good for Black. 9 Qb3 can be met with 9 . a5 10 a3 a 4! 1 1 Qxb4 ( 1 1 Qd1? Bxc3+ as after 9 a3) l l . . . Nc 6 1 2 Qc5 when, pace Burgess, 12 . . . Nd7? does not trap the queen and 13 Qb5 wins - instead 12 . . . Ra5 13 Nb5 Nd7 14 Nxc7+ Qxc7 15 Qd6 Qb6 re­ sults in an untested position which looks reasonable for Black. .

9

.

...

c5 (Diagram 1 3)

Diagram

13

An attempt to justify the move order

Diagram

14

An interesting tactic

This is the point of Black's play.

10 Be2 White does best just to develop; neither 10 a3 cxd4 1 1 Qxd4 Qxd4 1 2 Bxd4 Bxc3+ 1 3 Bxc3 Na4 1 4 Bb4 !? Nxb2 1 5 Nd4 a 5 nor 1 0 Qb3 cxd4 1 1 Nxd4 Qh4+ 12 g3 Qe4 13 Kf2 Bxc3 14 Bg2 Qxd4 (a very strong queen sacrifice) is desirable.

1 38

F ou r P a w n s A tta c k

1 0 ... Nc6 1 1 0-0 cxd4 I t isn't clear e i t h e r t h a t 12 d5 w a s a threat (both 12 . . . Bxc3 and 12 ... Na5 are conceivable i n reply) or t h a t 12 dxc5 i s a good idea, and if not i t m u s t be better for Black to keep t h e tension, so 1 1 . . . 0 - 0 might be better. It has n't been played more tha n a couple o f times, s o I have no theory to give to you, but 1 2 dxc5 Nd7 1 3 Na4 Nxe5 1 4 Nxe5 Nxe5 15 Qb3 Bg4 ! is a n idea worth knowing.

1 2 Nxd4 Nxd4 13 Bxd4 B c 2 (D iagram 14)

A slightly strange-looking move whose pu rpose is to force the queen to d2 a n d thus avoid 1 3 . . . 0-0 1 4 N b 5 ! etc.

14 Q d 2 Rc8 1 5 Kh1 1 5 Rf4 i s also dangerous,

just g o i n g for

it on the kingside.

1 5 . . .0-0 15 . . . Nxc4? drops a piece to 1 6 Qxc2 Qxd4 1 7 Q a 4 + .

1 6 b 3 Bc5 17 Nb5 Bxd4 18 Nxd4 B g6 It is high time to get this piece 'off prise' before an accident h a ppens, for e x a m p l e 18 ... a 6 19 Bf3 Rc7 20 Qe3 and Nxe6 after the bishop moves.

19 Bf3 Rc7 20 Qe3 Rd7 2 1 Rad 1 a 6 (Diagram 1 5)

Diagram 1 5 H o w might White continue?

Diagram 1 6 White is in the d riving seat

22 R d 2 Up to here the g a m e had fol lowed a n earlier contest between Craig Pritchett a n d Howard W i l l ia m s - the spiritual father of the 8 . . . Bb4 variation - w h i c h was considered to demonstrate equ ality for Black a n d in which White h a d continued 22 Be4. I don't u n derstand that move at a l l - i n fact I w o u l d say that it was a positional error. What White w o u l d like i n this position is t o u s e his space adva n tage to dominate the d-file, i f possible, and to press for mate on the dark squares o n t h e kings ide if the opportu nity arises. I f h e h a s to ex-

1 39

S t a rt i n g O ut : A l e k h i n e ' s Defe nce

change a n y m i nor pi eces at all, then the bes t trade for White would be h i s bishop for Black's k n i g h t , which can defend the d a r k s q ua r e s . 2 2 Be4 p rom otes n o n e o f those a i m s . This pawn s tructure is deceptive. It looks very nice for Black, but the extra s pace it gives White i n the m iddlega m e , a n d his pos s ibilities o f a kings ide attack, s h ould not be underes tim a t e d . In fact Black often needs to

break with . . . f7- f6/f5 i f

he c a n .

2 2 . . .Q c 7 2 3 h 4 h 6 2 4 h 5 Bh 7 2 5 Bg4 The teasing threat o f 26 Bxe6 fxe 6 27 Rxf8+ and Nxe6+ ham pers Black in

contesting the d-file. I t seem s t o m e t h a t White h a s all the

chances here, and t h a t Black ought to look into 1 1 . . .0 - 0 .

2 5 ... R e 7 2 6 Rfdl N d 7 2 7 R e l White h a s n't really lost tim e - t h e knight i s wandering around l i k e a lost soul a n d on d 7

prevents t h e rooks from challenging the d-file. I n

fa ct i n Pritchett-Will i a m s the kn ight went to c8, a n d I wonder i f t h a t was n't bes t here , t o o .

2 7 ... Nc5 2 8 Nf3 N e 4 2 9 Rd4 N g 3+ 3 0 Kh 2 Nf5 31 Bx£5 Bxf5 This

exchange brings Black n o relief. T h e knight w a s n o t on a good

circuit, but at least it was c a p a b l e of d e fe n di n g the dark squares.

3 2 g4 Bh 7 3 3 Rd6 (D iagram 1 6) White dom inates the board. This rook ca nnot be c h a lle n g ed without a n enorm ous pa s s ed pawn arriving o n d6 i n its pla c e

.

3 3 .. .Q a 5 34 Re2 Q a 3 3 5 R d 2 Qc l This queen tour is a gra ph ic i l l ustration of Black's suffering.

3 6 g5 Q b 1 36

. . .

hxg5 37 Qxg5

Ree8 38 h 6 Bg6 39 hxg7 Kxg7 40 Qf6+ Kg8 4 1

R6d4 followed by Rh 4 is n o

better. Black's h e l p less ness i s striking.

3 7 gxh 6 Q£5 38 hxg7+ Qxh5+ 39 Kg1 Re S 4 0 R 6 d 4 f5 41 exf6 1-0 A very instructive game i n which Black's mistakes are by no m eans easy t o p i n po i nt.

Stati sti cs Black has scored 5 1 % after 8 . . . B b 4 .

Theoretica l ? going t o s u rvive t h e opening a s long a s he variation, though, c a r e · fu l attention to the early m ov e order is essential. If Black all ows Bd3 to h a p p e n in one of the 'bad' pos itions for him, he r a t e s to get mauled.

Not terribly. Black i s

knows up to 9 . . . c 5 . If you do w a n t to play this

Theoretica l Status H a s a high reputation, but I don't l i k e t h e g a me l i n e a t a l l . I would look into 1 1 . . . 0 - 0 i f I wanted to p lay this lin e , and i n a n y cas e if this i s

1 40

F o u r P a w n s A t ta c k

to be a sta nd-alone system 7 Nf3 N a 6 8 exd6 also needs to be made t o work. I don't see t h i s happening.

Game 33 0 S .Movsessian • Luther Istanbul 2003 1

e4 N f6 2 e5 Nd5 3 d4 d6 4 c4 Nb6 5 f4 dxe5 6 fxe5 c5 (Diagram

1 7)

Diagram

Diagram

17

U ltra-provocative play

18

Onward !

A move with a curious hi story and wild by-ways. First investigated i n the 1 920s, for Eales a n d W illiams i n 1 97 3 it h a d ' n o redeeming fea ­ tures' yet today i t i s quite a respectable alternative to t h e m a i n lines,

although there remain many complicated a n d u nsolved questions .

7 d5 The apparently feeble 7 Nf3 cxd4 8 Qxd4 Qxd4 9 Nxd4 actually has an idea behind it - the u nwary 9 ... Nc6 is m e t by 10 Nb5, whic h causes difficulties. I am not convinced by the usu al recommendation of 9 . . . e6 10 N b 5 N a 6 1 1 Be3 Bb4+ 12 Nd2 0 - 0 either: it seems to m e that W h ite is better after 1 3 a3 (and not 13 Nxa7? w hic h merely develops Black's pieces). I t h i n k Black ought t o look into the untested 7 Bg4 8 . . .

d5 N 8 d 7 . 7 dxc 5 ? ! , on t h e other

hand,

is good for

Black: 7

. . .

Qxd l + 8

Kxd l N a 4 9 b 4 a5 10 a3? axb4 1 1 axb 4 Nc3+! etc. Of course Wh ite doesn't have to fall for th is, but if h e can't maintain his queenside pawn wedge he's simply worse. 7

.••

e6

The original idea a n d still the main line, b u t 7 . . . g 6 i s a n alternative syst e m . I t has never caught o n at a high level a n d is not well ex­ plored, so I haven't given too much detail, although it is worth noting that Black has scored 5 7 % . Black intends simply to develop with ... Bg7, . . . 0-0 and . . . Bg4 t o target the e 5 - p a w n . Normally h e

plays

. .

.e 7 -

e 6 also, a n d W hite h a s t o b e very c a reful about d 5 - d 6 i n reply, whi c h

141

S ta rt i ng O u t: A lekhi ne 's Defe nce can lead his centre

into an elephant trap, on e game, for e x a m p l e , con­ 8 Bf4 Bg7 9 Nc3 0-0 10 Bd3?! (an ineffective square) 10 . . . e 6 1 1 d 6 Nc6 1 2 Nf3 N d 7 1 3 Qe2 Nd4 1 4 Nxd4 cxd4 1 5 N b 5 Q a 5 + 1 6 Bd2 Qb6 17 b4 Nxe5. The other thing for White to avoid i s accidenta lly l os ing the e5-pawn to s o m e combination of . . . Bxf3 a nd . . . N 8d7, for ex­ a mpl e (instead of 10 Bd3 , above) 10 Be2 e 6 1 1 Nf3 exd5 12 cxd5 Bg4 tinuing

1 3 0-0 N8d7 14 Qe 1 Bxf3 1 5 Bxf3 Nxe5 ! 16 Bxe5 Re8 was another ga m e . The main line is considered to be 8 Bf4 Bg7 9 N c 3 0 - 0 10 Qd2 e6 1 1 0-0-0 exd5 1 2 cxd5 Bg4 13 Rel c4 14 h 3 Bf5 , p l a n n i n g a pawn

sacrifice with . . . Bd3. The spectacular ga mes Volzhin-L.Sveshnikov, 1 988 (a n n o ta t e d

on

Ch e s s base a nd

in Davies)

and I l incic-Marin kovic,

1 989 (a nnotated in Burgess) were what launched this variation into the

li mited popularity it has e nj oyed

since.

Perh aps White should reflect that if he can m a in ta in the pawns at e5 a n d d 5 he i s bound to stand well, a n d t o do that w h a t he needs is his

knights undisturbed on c 3 a nd f3 and his bishop on f4. He does not

particularly need his ki n g s bishop, and what he certainly doesn't '

need is to ca stle q u e e n si de into the line of Black's fianchettoed bishop

and i n front of Black's knights, thus prese nting Black with counter­ play. So I liked White's pl a y

in Novak-Havas,

Pula 2000, which went

8 Bf4 Bg7 9 h 3 0 - 0 10 Nf3 e6 1 1 Nc3 exd5 (this might be premature in view of White's 1 3th) 12 cxd5 N8d7 13 Bb 5 . Black is already i n diffi­ cu lty here, and the game went 1 3 . . . a6 1 4 Bxd7 (of course) 14 . . . Nxd7 15 0-0 c 4 1 6 Qe2 N b 6 1 7 Bg5 f 6 1 8 exf6 Bxf6 19 Bxf6 Rxf6 20 Nd4 a n d W hite won han dily. If I w a s going to play

this system t h i s i s

what

would trouble me.

8 Nc3 8 d 6 loses to 8 . . . Q h 4+ 9 g3 Qe4+, but I a m a fraid a long variation

(which dates from the 1 920's!) is needed to prove it since White can i n fact trap the queen : 10 Q e 2 Qxh l 1 1 N f 3 N c6 1 2 N b d 2 N d 7 1 3 Kf2 N dxe5 1 4 Nx e5 Qxh2+ 1 5 Bg2 N d 4 1 6 Qdl Bxd6 1 7 Nfl (at last!) 17 . . . Qxg2+ 18 Kxg2 Bxe 5 .

8 . . . exd 5 9 cxd5 c4! (Diagram 1 8) the key move of the line. The p a w n is not e n prise because of Qa4+) . Nonetheless the move does a mount to a pawn sacrific e , the idea b e i n g that Black wants to d ev e l o p h is kin g s bish o p actively and is prepared to give up both a This is

10 Bxc4?? Qh 4 + (and not 10 . . . Nxc4?? 1 1 '

tempo a n d a pawn to do i t .

10 d6 'l'his move leads to razor-sharp com plications. The ' m a in lin e ' is 1 0 Nf3. No w the older 1 0 . . . Bg4 was played i n t h e wonderful g a m e Bron­ stei n - Ljuboj evic, Petropolis Interzonal 1973, which has been a n n o ­ t a t e d i n many places, notably by Speelman in Best Chess Games 1 9 70- 1 980 a n d T im m a n i n Art of Chess Analysis (both long out of pri n t b u t must- buys if you c a n fi n d them) . The verdict is that White stands well in some awesome

1 42

com plications starting with 1 1 Qd4

F o u r P a w n s Atta c k

Bxf3 1 2 gxf3 B b 4 1 3 Bxc4 0 - 0 14 Rg 1 , b u t si nce these are well covered elsewhere and i n any case the quiet 1 1 Be2 also seems to be good I haven't given the details. The modern line is 1 0 . . . Bb4 a n d the pawn s acrifice I referred to i s 1 1 Bxc4 Bxc3+ 1 2 b x c3 Nxc4 1 3 Qa4+ N d 7 1 4 Qxc4 N b 6 1 5

Qb5+ Q d 7 1 6

Qxd 7+ K x d 7 1 7 d 6 Rc8. Black h a s good compensation for the pawn i n this endgame, w h ich i s well a n alysed i n other books. 1 1 Bg5 Bxc3+ 1 2 bxc3 Qxd5 1 3 Qxd5 N x d 5 1 4 0-0-0 N e 7 1 5 Bxe7 K x e 7 16 Bxc4 B g 4 1 7 Rhf1 N c 6 i s a l s o nothi n g t o write h o m e a b o u t for White. 1 0 a 3 ? ! Bc5 prevents kingside c a s tl in g and is fine for Blac k ; lovers of

one - movers s h o u l d immediately get out their databases and look up Black's twenti eth move i n Vete m a a - Shabalov, USSR 1 986. White's other purposeful move is 10 Be3 Bb4 1 1 Bxb6, which allows him to m a i n t a i n t h e d 5 - pa w n , but Black can develop all his pieces with 1 1 . .. Qxb6 followed by . . . 0-0, . . . Bg4, . . . N d 7 and so on and should have no problems. Other moves tend to allow Black's counterplay to re­ move the c e n tre, for e x a m p l e 1 0 Qd4 Nc6! 1 1 Qe4 Nb4, threatening both the d - p a w n a nd t h e e m b a rrassing . . . g 7 -g6, or 10 Bf4 Bb4 1 1 Bxc4 Nxc4 12

Q a4+ Nc6 1 3 dxc6 Nxb2 a n d Black w i n s .

T h e text m o v e i s by far Wh ite's shar pest a n d indeed h i s on ly critical test of Black's 1 i n e . He h e m s i n the bishop a t the cost of releasing c6 for the knight, and i n troduces the idea of Nb5-c 7 . The traditional re­ p l y is 10 . . . Nc6 1 1 Nf3 ( 1 1 Bf 4 g5 1 2 Ne4 gxf 4 is a similar queen sacri­

fice) 1 l . . . B g 4 12 Bf4 g 5 1 3 Ne4 gxf4 14 Nf6+ Qxf6 15 exf6 0-0-0. This w a s thought good for B lack for a while until Bur gess' 1 996 b o o k raised doubts a fter 1 6 Qcl . T h a t occurred i n the entertaining game Bryson - Luther, B l e d Olym p i a d 2002. Black had d r a w n on th a t occa ­ sion but computers will tell you t h a t White was w i n n i ng. I imagine both Movsessian and Luther k n e w this a n d i f the i nterested reader cares to feed Bryson-Luther to a computer i t w i l l co n fi r m h o w . G iven that 1 0 . . . Nc6 1 1 N b 5 Qh4+ 12 g3 Qe4 1 3 Kf2 Qxh 1 1 4 Bg5 i s also ex­ tremely da ngerous for Black ( B e n d e r-Rogulj, 2002 went 1 4 . . . f6 15 Nc7+ Kf7 16 e6+ Kg6 1 7 Be3 when 17 .. . f5 is unclear according to a comb ination of Fritz, Burgess and Flear, although I can't see Black's position catching on), I think the future lies i n Black neither tossing material nor com m i t t i n g his k i n g to g6

just yet.

IO B e 6 ! ? ...

Variously s u ggested b u t untested before this g a m e . Black's idea is to meet 1 1 Nb5 with 1 l . .. N d 5 (the . . . Qh4+ tl·ick is still d e fen di n g the c ­ p a w n ) a n d now 12 N f 3 Q a 5 + (and n ot Martin's 1 2 . . . N a 6 w h i ch basically loses to 1 3 Bxc4 Ne3? 14 Qa4). Black does h ave some other ideas. In h i s notes o n chesspublishing.com An d r e w Martin suggests 10 . . . g 6 1 1 Nf3 Bg7 b u t now 1 2 Bg5 f6 1 3 exf6 Bxf6 14 Bxf6 Qxf6 1 5 Bxc4 Nxc4 1 6 Qa4+, which i s supposed to be good for White (it certainly looks it), i s considered the refutation, a n d h e doesn't mention this.

1 43

S t a rt i n g O u t : A l e kh i n e 's D e fe n ce

1 1 Nf3 Nc6 1 2 Nb5

Martin gave 12 Be2 g6 as satisfactory for Black, but he doesn't say what he's going to do after 1 3 Bg5, and to be fra n k I don't have any idea myself: 13 ... Qd7 14 Bf6 Rg8 15 Ng5 looks extremely strong, and 1 3 . . . Qc8 14 Nb5 even worse. I think Luther's intention against both 12 Bf4 and 12 Be2 was more likely 1 2 ... N d 7, creating a decent post for the queen on a5 a nd exerting pressure on e 5 . In due course Black can break up the e5/d6 pawn duo with f7 -f6. ...

TIP: Once W h i t e g e t s this e5/d6 structure in t h e Fou r Pawns Attack Black a l m ost a lways needs (i) to p lay ... Nd7, ... Nc6 to attack eS, ( i i ) to keep the poss i b i lity of ... f7-f6 and (ill) t o watch out for Bg S!

12 ReS 13 Bg5 ...

The bishop isn't so great here, but I think Movsessian wanted to stop the above-mentioned plan before the knight can move to d7 and open the path to a 5 for the queen. 13 Nxc7+ Rxc7 14 dxc7 Qxc7 is good for Black - h e has ended u p tossing material after all but at least it wasn't a queen, and his active pieces and White's lagging develop ­ m e n t and unsafe king ( t h e bishop w i l l come to c5) give Black the a d vantage. ­

1 3 ... Q d 7 1 4 a4 (Diagram 1 9)

Threatening to win the c4- pawn after a 5 . 1 4 . ..h 6 1 5 Be3 N d 5 1 6 B f2 a 6 1 7 N b d 4

1 7 Bxc4 axb5 1 8 Bxd5 i s met n o t b y the tempting 1 8 . . . Bxd5 1 9 Qxd5 Nb4, since White has good co mpensation for the exchange after 20 Qxb5 Nc2+ 2 1 Ke2 Nxa 1 22 Rxa l Qxb5+ 23 axb5 f6 24 Rd1 Kd7, but 18 . . . Nxe 5 ! 1 9 Nxe5 Qxd6, recovering the piece with a good game. As Martin says, after 17 Nbd4 a critical position has been reached. Black needs to act before the c - pawn simply drops.

Diagram

19

White's aggression continues

1 44

Diagram

20

Drastic action

F o u r P a w n s A tta c k

1 7 ... Nxe5!? It's not clear - well it's not clear to me, at least - that this was actu­ ally necessary. Black could have continued 1 7 . . . Nxd4 18 Qxd4 Nb4 19 Qd2 Nc6, when the c-pawn is defended and Black can go o n with ... g7g5, ... Bg7 and 0-0. At least a l l his pieces are playing and he's not mater ial down. But the text i s a reasonable gamble over the board. . ..

18 Nxe5 Qxd6 19 Nef3 Qf4 (Diagram 2 0 ) Blac k h a s two pawns for t h e piece, w ith active forces, a n d t h e threat Bb4+ is going to force White's king to move . O n the other hand White i s in no danger of getting m ated quickly, the pawns themselves are n o great threat, and White has no weak nesses. The position i s un­ clear - perhaps you'd want to be Black over the board and White in correspondence, as i s so often the case with these speculative sacri­ fices. of

. . .

2 0 Be2 Bb4+ 2 1 Kf1 0-0 2 2 Nxe6 The crucial position for assessing the sacrifice. Martin suggests 22 Qc 1 Qf6 23 Nc2 Bd6 24 Bd4, planning 25 Kf2, but Black is still pretty active - even computers give h i m equality and over t h e board he m u s t have good chances. But the text does seem to be good if followed up correctly.

22 ... fxe 6 23 Qd4 Qxd4 24 Bxd4 Black looks to be i n trouble, but h e has a shot.

24 ... e5! (Diagram 2 1 )

Diagram 2 1

Diagram 22

A necessary sacrifice

Black maintains the pressure

25 Bxe5 N e 3+ 2 6 Kgl Nc2 27 Rc l Fritz prefers 27 R d 1 with the idea that 2 7 . . . Bc5+ 28 Kf1 Ne3+ 29 K e 1 N x d 1 3 0 Bxc4+ Kh8 3 1 Kxd 1 i s good for White. I m ust a d m i t I can't see why Movsessian rejected this - possibly he didn't see the game continuation a nd thought he was just winning in the game, o r maybe time trouble was alrea dy a factor.

27 . . . Bc5+ 28 Kfl Ne3+ 29 Ke l Nxg2+ 30 K d l N e 3+ 3 1 K d 2 Red S+

145

S t a rt i n g O u t : A l e k h i ne 's Defe n c e

(D iagram 2 2 ) 3 2 Kc3? The o n ly move according to Martin, although to m e it seems to be a losi ng blunder. 32 Kel may be met by 32 . . . Ng2+ 33 Kfl Nh4, although whether Black has a decisive continuation after 34 Rc3 isn't clear. I n fact since rushing in with 3 4 . . . R d 3 3 5 B x d 3 Rxf3+ 3 6 K e 2 cxd3+ 37 Rxd3 certainly isn't the answer, a nd meanwhile White threatens j ust to take the c-pawn with check a nd continue with Be2, Bg3 and so on, I'm not sure how Black continues.

3 2 ... Rx£3! Ouc h . Movsessian cannot have seen this move. He reacts well now; the play e rs must have been in desperate time trouble. O f course 33 Bxf3 Rd3 is mate.

33 b 4 ! ? A fine swindling attempt, b u t Black is s t i l l winning.

3 3 ... Nd5+ 3 4 Kb2 Rb3+ 35 K a l Ra3+ 36 K b 2 Rb3+ 3 7 Kal Ra3+ 38 Kb2 lf2-'h A very hard decision to understand on Black's part; the obvious 35/37 . . . Rxb4 would have left him with a winning position, three pawns up, a n d he has a large number of other winning moves too.

Statistics Black has scored a respectable 47%.

Theoretical? Yes and no. To be honest, the statistics don't count for much, since i n s o m a ny o f t h e 400 o r s o games i t h a s been clear that o n e side o r the other has n o idea what they are doing and has simply lost without a fight. O n the other hand there isn't much theory a fter 1 0 d6 Be6. Black coul d get by with knowing that move and the sequence lea d ing to the ending after 10 Nf3 Bb4 , and working the rest out.

Theoretical Status Now you're a sking! 10 d6 is the only critical move. I n the end I don't think the queen sacrifices are going to prove viable, fun though they are. I predict that theory will develop with 10 ... Be6, and given that there's only one game, there isn't really any 'theoretical status' to tell you about. All we can say is that Black's idea made a successful debut. Game 34 0 Kr a nzl • Ba u mga rdt ICCF e-mail 2000

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 c4 Nb6 5 f4 g5!? (D iagram 2 3) You have to love this move. As Kasparov said, chess is a n in teresting game if such moves are possible. Whether it's any good is another

146

F o u r P a w n s At ta c k

matter, b u t before we get o n t o that, I prom ised y o u a quick look a t 5 . . . g 6 a n d Wahl's 5 . . . c5 gambit.

D iagra m 2 3

Diagram 2 4

H m m m. . .

B l ac k is struggling

The idea of 5 . . . c5 is to meet 6 dxc5 (the only sensible move as 6 d5 e6 is fine for Black) not with 6 . . . dxc5 7 Qxd8+, which is not much fun , b u t 6 . . . N 6 d7 7 c x d 6 exd6 8 e x d 6 Nf6. If White continues m undanely now w i t h 9 Nf3 Bxd6 10 Be2 0-0 then Black i s certainly going to ob­ tain very adequate compensation with a continuation like 1 1 0-0 Bc5+ 12 K h 1 Qxd1 1 3 Bxd 1 Ne4, but i n s t e a d of such feebleness White has a number of dangerous ideas. O n e is 9 c5!? with the idea 9 . . . Qa5+ 1 0 N c 3 Qxc5 1 1 Qe2+ B e 6 1 2 N b 5 Na6 1 3 a 3 , maintaining t h e d6-pawn and pl a nn i ng Be3 a n d Rd1, when Bl ack will struggle to s ta y afloat. However, the plan which has caught White's eye in encounters i n this variation so far i s 9 Qe2+ Be6 1 0 Nc3 (not 1 0 f5? Q a 5+ but now this is a threat) 10 . . . g6 1 1 Be3 B g 7 (at least Black ought to capture the d­ pawn here) 12 0-0-0 0 - 0 1 3 N f3 Nc6, when Black doesn't have value for his pawns. I don't think this gambit has too much of a future from a theoretical point of view, a lth o ugh it could give a n unwary opponent a nasty shock . 5 . . . g6 is a completely different matter. Note that 5 . . . dxe5 6 fxe5 g6 is a hybrid of this a n d the system i n the note to Black's seventh in Game 33. Black decides to keep . . . c7-c5 for later. I t seems to m e that White could meet that with 7 Nc3 Bg7 8 Bf4 0-0 9 h 3 as I reco mmended in Gam e 33, but this i s ano ther largely untried line. By con t r a s t , in the present sy s t e m White doesn't have f4 for his bishop, but on the other hand if Black isn't careful dxe5 will become a strong r e p l y to . . . dx e 5 . White hasn't unearthed a convincing reply to 5 . . . g6 at the moment. Davies has suggested 6 Nf3, de v elo pin g the kingside a nd reserving Nbd2 to defend the c - pawn (which doesn't look terribly threatened at the moment, I know, but bear with me) . Black might consider meeting that with 6 . . . dxe5 7 fxe5 Bg4, but the position i s u n explored. Instead

147

Starting O ut: A lek hin e 's Defe nce

the m a i n line, insofar as there is one in such a backwater, is 6 Nc3 B g 7 7 Be 3 , and now Black has b e e n doi ng w e l l with the cunning 7 . . . Be6. The idea of Black's move order is to by p a s s the supposed refu­ tation 7 . . . 0-0 8 c5 N6d7, when 9 h4 is very dangerous, by defending d 5 in order to allow 8 c5 N d 5 . Returning to 7 . . . Be6, 8 Qb3 a5 s ee m s to force 9 a4 when . . . Na6-b4 gives Black very decent play. Nobody seems that wo rri e d about 8 d5, which Black can meet with an . . . e 7-e6 plan, for e x a m p le 8 . . . Bf5 9 Be2 0-0 10 N f3 e6. This lea ve s 8 Nf3 0-0 (the c4p a w n was i ndirectly defended b y the usual Q a 4 + , but n o w it is really en prise), when the best White has m a na ge d i s 9 Qb3 a5 10 Ng5 ( 1 0 a 4 Na6-b4 again) 1 0 ... a 4 1 1 Nxe6 axb3 1 2 Nxd8 Rxd8 1 3 a 3 Nc6, which produces a fairly equal ending. 9 Nd2 fxe5 1 0 dxe5 c5 is worse, and 9 b3 c5 10 dxc5 N 6 d 7! (a la Wah l ! ) 1 1 cxd6 e x d 6 12 Qxd6? Nx e 5 is worse still. 5 . .. g6 evidently deserves more popularity. It looks s l i ght ly strange to d e v e l op t h e king's bishop o n g7 when t h e whole point of White's play h a s b e e n to fortify e5, but keeping White's pawn o n ready

f4 until Black is

for . . . dxe5 a n d . . . c7 -c5 has obvious merits.

Anyway, re t u r ni n g to the position after 5 . . . g5, the poi n t of this m o ve

is clear. Black intends to remove the f-p awn a n d thus dismantle White's pawn centre . A good many books on the Alekhi n e have dis­ missed the move with a '?' - Da vies isn't so sure, and I'm goi n g to sit on the fence and say that I think it's probably bad a t the e n d of the day, but it's not t h a t s i m p l e . Two things are clear: most of the pub­ lished refutations are rubbish, a n d the statistics say 5 . . . g5 is a good move; Black has scored 59% on my database, a l thou gh in la r ge part that is due to the efforts of Michael Schirmer, a German correspon ­ dence player whose baby this line is, a nd who has s core d a s te l l a r 22/30 with it.

6 exd6 A very logical move, since it gets the e5-target out o f the way before

taking on g5 . The alternatives are Nf3, Nc3, Qh5 and fxg5 ( 6 d 5 e6 is no good) . 6 fx g 5 is perhaps t h e least effective of these, but it does pro v i de a good illustration of Black's idea with 5 . . . g5. Black replies 6 . . . dxe 5 , when White is forced into 7 d5 (7 dxe5 Qxd1 + 8 Kxd 1 Bg7 9 Nf3 Bg4 is aw­ ful) a n d after 7 ... e6 Black already stands better. 6 Nf3 rather a s ks for 6 . . . g4, which is Schirmer's choice and Davies' rec· o m m en da t i on . 7 Nfd2 dxe 5 8 dxe 5 Q d4 is a l ready better for B lack , but 7 Ng5 is not so clear - presumably Black p l ay s 7 . . . dxe5 and now rather than 8 fxe5 Nc6 White has to play 8 c5 Nd5 and now 9 dxe 5 . This is j u s t a little irritating for Black as 9 . . . e6 leaves g4 en prise, 9 . . . Nc6 1 0 Bc4 is a n noy i n g and 9 . . . Bf5 10 Nc3 e6 1 1 h3 likewise (l l . . . Bxc5? 1 2 hxg4 Bg6 1 3 f5 ) . I ' m not su re Black isn't better off with the a rgua b ly more thematic 6 . . . gxf4 7 Bxf4 Bg7 8 Nc 3 Nc6 9 Be2 dxe5 1 0 dxe5 Bg4 , when he has all hi s pieces active and s ee m s to stand we ll .

1 48

F o u r P a w n s Attack

6 Nc3 is also sensible. I t i s not possible t o transpose to the above l i n e s i n c e 6 ... gxf4 7 Bxi4 N c 6 (or 7 Bg7) 8 c 5 i s good for White. Instead Schirmer has p re fe r r e d 6 ... e6 to bolster the d5-square, a l th o ugh 7 exd6 cxd6 8 fxg5 then looks very strong to me. If this is the best Black can d o then 6 Nc3 is a serious probl e m . ...

6 Q h 5 w a s given as a refutation b y Eales a n d Williams, a n d has a n intimidating look, but in fact Schirmer h a s proved that Black i s fi n e . I n r e p l y to 6 dxe5 7 Nf3 is t h e o n l y pressing move. A t o n e time 7 c 5 Nd5 8 fxe 5 w a s thought to be good, but Burge ss' 8 . . . Nb4 is virtually winning for Black - since the d-pawn cannot be defended (9 Q d 1 ? Q x d 4 a n d 1 0 . . . Nc2+) White i s obliged either to give u p a rook for a cou ple of checks after 9 Nf3 Nc2+ 10 Kf2 Nxa 1 1 1 Bc4 Qd7 1 2 Qxf7+ Kd8 and so on, or grovel with something like 9 Na3, le a vi n g the d ­ pawn to its fate. 6 Qh5 dxe5 7 Nf3 at first s i gh t looks very strong, but Schirmer's 7 . . . Bf5 gives Black good play, for example 8 Nxe5 Bg6 9 Nxg6 fxg6, or 8 fxe5 Nc6 9 Be3 h6 1 0 Nxg5? Bg6 1 1 Q h 4 Nxd4 1 2 Bxd4 hxg5 1 3 Qxh8 Qxd4 with more t h a n enough com pensation. ...

6 ... Qxd6 7 c5 Q e 6+ 8 Be2 White's line has some big names behind it - Hart declared that after 8 Qe2 here White was winni ng! This is an amazing misjudgment for such a player to make, as not only is White probably al re a dy worse a f ter that move, but also it is obvious that 8 Qe2 i s the worst candidate move at White's disposal since it allows Black to trade queens and thus free h is queen's bishop for action. I n fact the entire point of the otherwise hideously anti- positional c4-c5 was to force the queen to the horrid e6-square. Mter 8 Qe2 Nd5 9 fxg5 Bg7 10 Nf3 Nc6 1 1 Qxe6 Bxe6 1 2 Bb5 Ndb4 Black has excellent play and stands much better. Better is 8 Kf2 Nd5 9 Bc4 gxi4 10 Qf3 c 6 1 1 Bxf4 Qg6 (Davies; 1 l . . . Bg7 is Mi s c h k e Schi r m e r a n notated in Davies a n d on Chessbase by Schirmer) with unclear play. ­

-

,

8 Be2, however, is the most obvious m o v e and in my opinion poses a severe threat to the survival of Black's system .

8 Nd5 9 fxg5 Bg7 1 0 Nf3 Nc6 1 1 Nc3 0-0 1 2 Q d 2 ! ( D i a gram 2 4) . . .

Th is is the star move. Afte r 12 0-0? Ne3 Black is fine, but with this ruled out he is really struggling for counterplay. Black makes a logi­ cal atte m p t in this game, but comes off worse, and it must be an open question whether there is actually anything good he can do.

1 2 ... Nxc3 1 3 b xc 3 Q d 5 14 0-0 e 5 This w a s Black's idea, b u t White h a s the answers.

15 B b 2 exd4 16 c xd4 Bg4 17 Bd1! Another excellent move - the bishop takes part in the defence of d4 by coming to b3 or a4. For the moment Bb3 followed by d4-d5 is threatened.

1 7 . . .Qd 7 1 7 . . .Bxf3 18 Bxf3 Bxd4+ 1 9 Bxd4 Qxd4+ 20 Qxd4 Nxd4 2 1 Bxb7 is a highly u npleasant ending for Black.

1 49

Start i ng Out: A l ek h ine's Defe nce

18 Ba4 Rad8 19 Khl And that's it. Since 19 ...Bxf3 20 Rxf3 Bxd4 21 Bxc6 bxc6 22 Rd3 wins a p iece, Black has nowhere to go . 19 . . . Rfe8 20 Qf2 1 -0 I don't s uppose it was obligatory to resign, but White has consolida ted and is certainly winning. It was rather rude of me no t to give a game of Schirmer's, but the above was such a cle arcut performance it seemed imp oss ible to o mit it.

Statistics Black has scored 59 % after 5 ...g5, and 48% after 5 ...g 6. 5 ...c5 has only been pla yed in a handful of ga mes.

Theoretical? I n a word - no . By all me ans sit down and toss o ut 5 ...g5 - i f your op­ ponent isn't more or less on his own you'll be very unlucky. The s a m e goes for 5 ...g 6, altho ugh it's wise to h ave an ide a o f h o w Black's coun­ terpla y against c4 develops, and 5 ...c5.

Theoretical Status In my opinion, not so good for 5 ...g5. Both the game and 6 Nc3 pose Black a real and possibly ins oluble proble m. 5 ...g6, by contrast, is looking re as onable at the mo ment. 5 ...c5 is almo st certa inly unso und.

150

C ha pter Seven

T h e C h as e Va r i a t i o n

• •

I nt rod uction I l l u strative G ames

Starti n g O u t : Al e k h i n e 's D e fe n c e

I ntrod uction 1 e4 Nf6 2 e5 N d 5 3 c 4 N b 6 4 c5 N d 5 (D iagram 1 ) 4 c 5 really i s a n ugly move! W h e n w e learned the game w e were all taught t h a t one simply can't do this sort of thing to on e s l i ght squares and the d-file. White's intention, however, is to develop quickly and render such structural considerations irrelevant. He reasons that by bringing Black's knight back to d5 and cutting o ff its obvious retreat square he will be able to gain tempi on the knight to bring h is pieces into play quickly. Ideally he hopes to force . . Nxc3, when he will re­ capture with the d-pawn. He often leaves the c p aw n en prise to . . . Bxc 5 , which he will meet with Qg4 a nd the follow- up Bg5, perhaps forcing the sequence . . . Be7, Bxe 7 Qxe 7 , Qxg7. '

.

-

The Chase Variation, incidentally, is often used by players who also play the c3 Sicilian, since White more or less has to be w i l l i n g to transpose to a couple of variations of that opening. It is not theoreti­ cally dangerous but Black does well to have a line aga i n s t it in mind, a n d a n idea of his plans in that line, if only because the positions aris­ ing are a little unusual.

5 Nc3 White wants to play somet hing like the variation with 3 Nc3, but by moving his pawn to c 5 first he aims to prevent the 3 . . . Nxc3 4 dxc3 d 6 defence mentioned in Chapter 8. I n the present pos i tio n 5 . . . Nxc3 6 dxc3 d 6 7 cxd6! (this is the difference - White can relieve h i m s e l f of the doubled c-pawn) gives White a small a dv a n t a ge . It is al so possible to play 5 Bc4, when the next move will usually be 6 Nc3. 5 d4 is feeble as 5 . . . d 6 6 cxd6 exd6 (6 . . . cxd6 is a c3 Sicilian) is ea sy for Black. Meanwhile, fro m Black's point of view, White's structure cries out for . . . b7-b6. The exchange cxb6 a xb 6 obviously favours Black b e c au se h is pawns remain united while Wh ite's are split into two islands. The p l igh t of White's a-pawn is e s p e cia l l y sad , i s ol a te d on an open file if Black should continue with . . . Nxc3, bxc3. This also enables Bl ack to achieve the desirable exchange of light-squared bis hops w i th . . . Ba6 , working for his knight o n d5 and highlighting the weakness of White's light squares after his first four moves. Black has three main ways to meet the Chase Variation. 5 ... c6 with­ out ... e7-e6 is seen in Game 35, 5 . . . e6 6 Bc4 c6 in Game 36, and 5 . . . e6 without 6 . . . c6 i n Game 37.

I l l ustrative Games Game 35 0 T.Thorh a llson • G.Bu rg ess Gausdal 1997 1 e4 Nf6 2 e5 Nd5 3 c4 Nb6 4 c5 N d 5 (Diagram 1 )

1 52

Th e C h a s e V a ri at i o n

Diagram 1

Diagram 2

The knight is forced back to d5

Black seeks to clear the centre

5 Nc3 c6 6 Bc4 Usually White will meet 5 . . . c6 with both Nc3 and Bc4 on moves 5 a n d 6. 1f h e has started with 5 N c 3 then either 6 d4 or 6 N f3 c a n b e m e t by 6 ... d 6 comfortably enough for Black. If White starts with 5 Bc4 then 6 d4 d6 is s till nothing to worry about. Black should be aware th oug h that 5 Bc4 c6 6 Qe2 will prevent 6 . . . d6 a n d oblige h i m to continue with 6 e 6, ... b 7-b6, . . . Ba6 as i n G a m e 36, s i n ce after 6 ... d6? 7 exd6 the e-pawn is pin ned. ...

6 ... d6 (Diagram 2) With this variation Black is not ai ming to p u n ish White for his struc­ tural misdemeanours, rather to liqu idate the advanced pawns i m m e ­

diately a n d equalize. B u t White h a s saved a tempo by attacking d 5 twice before going d 2 - d 4 . He can n o w attack t h e knight a third time and simulta neously pin it.

7 Qb3 This is not the only move but it is the most testing a n d popular. White needs to take radical measures - if he lets Black develop and maintain the d 5 stronghold he can hope for equality at best. a) 7 Nxd5? is a mistake: 7 ... cxd5 8 Bxd5 e6 (and not 8 dxe5? 9 Qb3) 9 Bf3 dxe5 is already better for Black. ...

b) After 7 exd6 Black does best to flick in 7 ... Nxc3 (otherwise 7 . exd6 8 Nxd5 c x d5 9 Bxd5 dxc5 10 Qb3 is a sh ade inconvenient, although possibly nothing tragic) 8 dxc3 exd6 when again he is already equal . . .

c) 7 Qf3 is met - as in the main game - by 7 ... N d 7 ! , which is even stronger with the knight com ing to e5 rather than c5, hitting the queen and defending f7, e.g. 8 Bxd5 (8 exd6 misses the threat: 8 . . . Ne5! 9 Qe4 Nxc4 10 Qxc4 exd6 1 1 Nxd5 Be6 is good for Black) Nxe5 9 Bxf7+ Nxf7 10 d4 g6 1 1 cxd6 Qxd6 and Black is fine, the two bishops and develop­ ment compensating for the slightly ragged pawn structure.

1 53

S t a rt i n g O u t : A l e k h i n e ' s Defe n ce

d) The main alternative is t h e dull 7 cxd6 exd6 8 Nxd5 (or 8 Nf3 Be7 9 0-0 Nxc3 with equality) 8 ... cxd5 9 Bxd5 dxe5 10 Qb3 Qc7 1 1 Nf3 Bd6 12 0-0 0-0 which is equal, for example 13 d3 Nd7 14 Be3 Nf6 15 Rac 1 Qe7 1 6 Bg5 h 6 .

7 . . .N d 7 T h e k e y move of t h i s variation. Black seems t o b e i n trouble d u e to the pressure on d5 and f7 but he has some neat twists in mind. 7 ... dxe5 may also be playable but is tenuous.

8 Nxd 5 8 exd6? Nxc5 is bad, while 8 cxd6 Nxe5 is complicated and untried but is thought to be good for Black, a sample line being 9 Nxd5 cxd5 10 Bxd5 e6 1 1 Bxb7 Rb8 12 Q a 4+ Bd7 1 3 Qxa7 Rxb7 14 Qxb7 Bc6 1 5 d 7 + Ke7 1 6 Qb4+ Kxd7 1 7 Qd4+ Bd6 which i s terrible for White.

8 . . . Nxc5 The point .

9 N c 7 + (Diagram 3 )

Diagram

3

The pace has q uickly changed

Diagram

4

Black's king seems f i ne

This i s forced, otherwise White loses material, for example 9 Nf6+? exf6 1 0 Bxf7+?? Ke7 1 1 exf6+ gxf6.

9 . . . Kd 7 Black h a s a major alternative in 9 . . . Qxc7 10 Bxf7+ Kd8 1 1 Qe3 d5 (or 1 l . . .e6!?) . This is little played a n d theoretically stands well b ut it does not seem to be trusted by the gra n d m asters who play this l i ne. At first sight the position looks terrible for Black , b u t the point is that White's bishop is well -nigh trapped and if it is e xch a n ged for Black's knight White will be awfully weak on the l i g h t squares. White needs to avoid 1 2 Bh5? Qxe5, and 1 2 Nf3? Bf5 1 3 0-0 e6 is also feeble. The only GM ga me in this line continued 1 2 d4 Ne6 13 Nf3 ( 1 3 Bh5!?) 1 3 . . . g6 1 4 h4 Ng7 a n d now 1 5 h5 Nxh5 16 Rxh5 gxh5 1 7 e6 is reputed

1 54

T h e C ha se V a riat i o n

t o give White compensation.

10 Qe3 10 Nxa8? loses: Black takes the queen and mops u p the knight on a 8 later. 10 . Kxc7 ..

According to theory at the time of this game Black al ready stands bet­ ter. 11

Bx£7

The most natural move, although before this game it had been con­ sidered bad. 1 1 d4 Be6 1 2 Be2 Nd7 1 3 Nf3 Bd5 1 4 0-0 e6 probably does not give White enough for his paw n , while 12 b3 d 5 13 Be2 N d 7 1 4 f4 Bf5 i s better but s t i l l fine for Black. 1 1 . . . d5 1 2 Be2 Ne6 1 3 f4 g6 14 g4 h 5 - or here the untried 12 dxc 5 dx c 4 1 3 Bd2 Kb8 1 4 0-0-0 with a thoroughly unclear ga me - is also possible.

1 1 ...Qd7 In h i s book Burgess h a d given 1 1 . . . B f5 1 2 Bc4 d 5 13 Be2 d4, which indeed is horrid for White , but he no w avoids his own recommenda­ tion . I imagine the reason for this attack of shyness was 1 l . . .Bf5 12 d 4 when 1 2 Nd3+ 1 3 Kfl does not achieve a s much a s you might think after 1 3 . . . Nb4 i t seems to me that the chances are about e qu al The text is the prelude to a neat regrouping. . . .

.

1 2 d4 Q£5 13 Bc4 Be6 1 4 Be2 Nd 7 15 Bd 3 Q f7 1 6 N e 2 ! Better than 1 6 Nf3. The knight c a n take part in t h e fight for d 5 from c3 and also takes aim at f4. As usual in this system Black does not at all want to allow the exchange of his light-squared bishop for a knight.

16 . .. Nb6 1 7 0-0 Nd5 18 Qg3 g5 (D iagram 4) The position i s not cl e a r to my m i n d . Burgess gives Black the edge in NCO, but I would prefer to say that both s i de s have chances.

19 Be4 Rg8 20 Bd2 Bh 6 21 R a c 1 N f4 22 N c 3 Raf8?? A blunder. 2 2 . d 5 was necessary, with a very s harp position. . .

23 d 5! Bxd5 2 4 Bxd5 cxd5 White's point is that 24 ... Nxd5 25 exd6+ exd6 26 Nb5+ is devastating.

2 5 exd6+ Kb8 26 dxe7 Qxe7 27 Nd5 Qd 6 28 Bxf4 gxf4 29 Qf3 White is winning and t h e remainder is mopping up

29 ... Rg6 30 Kh 1 Rfg8 31 Rfe 1 a6 Not 3 1 . . .Rxg2?? 32 Qxg2 Rx g 2 33 Re8+; t h e insecure position of Black's king still tells.

32 Ne7 Rxg2 33 Nxg8 Rxg8 34 Rcd 1 Qb6 35 b3 Rc8 36 Re 2 R d 8 3 7 Rd e 1 Ka7 38 Re6 R d 6 3 9 Rxd6 Qxd6 4 0 R d 1 Qe6 4 1 Q d 5 1 -0

Statistics Good. Black has scored 59% w i t h t h i s variation, although from a

1 55

S t a rt i ng O u t: A l e k h i n e ' s D e fe n c e

small number of games. In fact the statistics are a poor guide throughout the Chase Variation; there are too few games a nd too many playable early options for them to mean much .

Theoretical? Very much so. There are some tactics Black simply must know, a n d there is a l o t t o learn for a fairly rare line.

Theoretical Statu s Unclear. There haven't been m a ny games between strong players ­ the fact is that gra ndmasters don't usually go i n for the Chase Varia­ tion. Game 3 6 0 H alser • Konopka Graz 2001

1 e4 Nf6 2 e5 Nd5 3 c4 Nb6 4 c5 Nd5 5 Nc3 c6 If B l a c k is aiming for this system it is probably better to start with 5 . . . c6, since 5 Nc3 c6 6 d 4 i s rare whereas 5 Nc3 e6 6 d4 is common. Probably it is fine to go 6 . . . c 6 even here, but White may be able to ex­ ploit the fact that h e hasn't gone 7 Bc4 to gain a shade of tempo once . . . Ba6 i s played, either by delaying a b i s hop move or p laying Be2/d3.

6 Bc4 e6 (D iagram 5)

Diagram 5 A strang e structure

Dia gra m 6 Black focuses on

d5,

White on e4

This is another variation which may horrify the casual onlooker. What about that horrible hole o n d6??? And surely Black isn't going to play the whole game with that cramping White pawn centre now that he's deprived hi mself of the . . . d6 break? It's not as bad a s all that . Black can remove the c-pawn with ... b 7-b6. H e will then either remove the e 5 -pawn too with ... f7-f6, or fight for

1 56

T h e C h a s e Va r iati o n

space on the kingside with . . .f7-f5 t o keep h im s e l f from getting mated. And he will also remove White's light-squared bishop with . . . Ba6, leav i n g himself with an inviolable outpost on d5. In fact this variation sometimes shows Alekh ine's Defence in its most classically hyper­ modern mould, as White's centre is dismantled from the side a n d Black's central p a w n s rule the board.

7 d4 This is t h e most popular move, but White has two other options. 7 Ne4 b6 8 Nd6+ Bxd6 9 cxd6 0-0 1 0 d4 Ba 6 1 1 b 3 c 5 i s fine for Black . I f

White ever wants t o castle h e can hardly avoid th e trade o f bishops, still less ca p t ure on d5, yet if he exchanges him self he can never hope to move Black's knight from d5. Meanwhile Black stands rea dy to dismantle the centre with .. .f7-f6. 7 Qg4 was p l ay e d b y the ever­ principled Russian GM Evgeny Sveshnikov on the only occasion h e faced thi s variation, a n d s o demands respect. Black should play 7 . . . f5 8 Qg3 b6 9 cxb6 axb6 1 0 Nge2 Ba6 1 1 d3 Qe7 1 2 0-0 Qf7 (the key rn a · noeuvre) , continuing his development with ... Be7 a n d ... 0 - 0 and e a r n ­ i n g a good game .

7 ...b 6 8 cxb6 a x b 6 9 N f3 9 N ge 2 is possible. U s u a l l y it i s b e t t e r for the white knight t o go t o e2 in this line in order to keep in touch with d5. O n the other h a n d after

9 Nge2 Ba6 1 0 Bb3 d6 Eduardas Rozentalis, one of the strongest p l a ye rs to use this line as White, was so disenamoured of his pros­ pects when faced with this position as to p la y 1 1 Bxd5 and offer a draw.

9 . . . B e 7 10 0-0 Ba6 1 1 Bxa6 Nxa6 1 2 Nd2 (D iagram 6) All according to the usual programme so far. Black is fine here. White now wants to take this knight to e4.

12 ... 0-0 13 Nde4 Nac7 14 Qd3 f5 1 5 exf6 Bxf6 16 Be3 Qe8 1 7 a 3 Be7 1 8 Rad l Q h 5 Black's play i s totally thematic. White has n o t hi n g constructive to undertake against Black's firm central pawn structure a n d the mon­ ster k n i gh t a n d Black has prospects of taking the initiative on the kingside . ,

1 9 B e l Rf7 20 Qc2 ( D iagram 7 ) 2 0 Q g 6 2 1 Rd3 Rf5 22 Rg3 Q f7 23 Rh3 Nxc3?! ...

T h i s seems a strange exchange. It i s true t h e recapture with the b­ pawn hardly ends White's structural problems (spare a thought for the poor a-pawn), a n d i n fact the c7 - knight was feeling a l i tt le redun­ dant, but White wasn't threatening anything and it was surely better to continue manoeuvring and prepare this move.

2 4 bxc3 h6 2 5 c4 Qf8 26 Rg3 Kh8 27 a4 Ne8 2 8 Rb3 Bd8 29 c5 bxc5 3 0 Nxc5 30 dxc5 was perhaps better. 30 ... Rxa4? 31 RbS is not possible, so t h e rook penetrates to b 7 and ties Black down. White h a s pressure after

1 57

S t a rt i n g O u t: A l e k h i n e ' s D e f e n c e

30 dxc5 Be7 3 1 Rb7 Rd5 32 Be3 Nf6 33 f3 ! , so Black must have gone wrong somewhere over the last few moves.

Diagram

7

Preparing to activate the rook

Diagram

8

An effective break

30 ... Nf6 3 1 R b 7 Rd5 32 Be3 Be7 33 N d 3 Bd6 34 h 3 Qe8 35 N b 2 e 5 ! (Diagram 8) Structurally Black doesn't really want to dissolve the d-pawn weak· ness, but this gives him a strong initiative.

36 dxe5 Qxe5 37 g3 Qe6 38 Kg2 Rh5 39 Rh1 Bxg3 ! 40 Bxh6 40 hxg3 Qxe3 or 40 Kxg3 Ne4+ 4 1 Kg2? Qg6+ 42 Kf1 N g 3 + were the points, although in the last line 4 1 K f3 Qf5+ 42 Ke2 Ng3+ 43 Kd2 Qa5+ 44 Kcl Nxhl 45 Bd4 still gives White chances of resistance (suddenly Black's pieces are not so well placed), and was surely better than the text.

40 ... Qd5+ 4 1 f3 4 1 Kxg3 Ne4+.

4 l ...gxh6 0-1 The rook comes to the g-file. A fine lunge i n time trouble. Although Black's earlier middlegame play wasn't too convincing, his opening was.

Statistics Good. Black has scored 53%.

Theoretical? No. Break with ... b 7 - b6, put the bishop on a6 and develop, adding .. . f6/f5 as necessary. The only manoeuvre to remember is .. .f5, . . . Qe 7, . . . Qf7 as a w a y of combating Qg4.

1 58

T h e C h a s e Va riati o n

Theoreti cal Status Good. White has so far shown nothing at a l l against t h i s li ne. Game 3 7 0 Roze n talis • Kuczynski Bank Pocztowy Open 2001 1

e4 Nf6 2 e5 N d 5 3 c4 N b 6 4 c5 N d 5 5 N c 3

White can also play 5 Bc4 first, of course . After 5 B c 4 e6 the main i n ­ dependent idea is 6 d4. 6 . . . d 6 n o w is a c3 Sicilian, o r Black can play

6 ... b 6 7 cxb6 axb6 and continue a s usual with ... Ba6 (a nd/or ... Nxc3 when the opportunity arises) . White occasionally tries 6 Qg4 when 6 ... Nb4 is a good reply. TI P: G e n erally in the A l e k h i n e it is always worth conside ring ... Nb4 i n reply when White plays Of3/0g4/0h5

5 ... e 6 (Di agram 9)

Diagram

Diagram

9

10

Black b eg i n s a worthy set-up

Black hits the c5-pawn

6 d4 White h a s an important alternative here in 6 Bc4 . This does of course leave the c-pawn en prise but White intends to meet 6 . . . Bxc5 with 7 d4 Bb4 8 Qg4. This i s complicated and little explored, a n d m a y well i n s o m e future world w h e r e chess is played entirely b y supercom puters be assessed as good for Black. Black has two easier alternatives, how­ ever, but both involve tricksome sequences and need to be learned. The first is 6 . . . Nxc3 7 dxc3 Nc6 (Black wants to te m p t 8 Nf3?! Bxc5, when White no longer has Qg4) 8 Bf4 Qh4. White's only way to defend the bishop without a l lo w in g 9 . . Nxe5 exploiting the skewer on the other one is 9 g3, when Black continues with 9 ... Qe7 and threatens to win the g-pawn with . . . g7 -g5. The most natural continuation is 1 0 Nf3 h6 1 1 Be3 b6 with a good game for Black (and not the instructive er.

1 59

St a rti n g O u t : A l e k h i n e ' s D e fe n c e

ror l l . . .g5? 1 2 Bb5, when White continues by capturing on c6; the re· this has on Black's queen's bishop i s striking).

strictive e ffec t

Alternatively, the lo n g - e s t a b li s h e d solution to this line is 6 . . . Nxc3 7 dxc3 Nc6 8 Bf4 Bxc5 9 Qg4 g5! 1 0 Bxg5 Rg8. T h i s leads to slight com· pl ications which are fine for Black .

6...b 6 6 . . . d 6 7 cx d 6 cxd6 i s a main line o f the c 3 Sicil i a n . B la ck can also play 6 . . . Nxc3 7 bxc3 b6 and indeed Nigel Davies advocated th a t in his book, although White can then cause difficulties (not necessarily i n ­ soluble ones) with either 8 Qf3 o r 8 Q g 4 (threatening Bg5). I t i s worth just noticing that 8 cxb6 axb6 9 Qg4 is ineffective as Black has 9 . . . c5, after which he can meet 1 0 Bg 5 with 10 . . . Qc7 - White's need to play Qg4 before cxb6 is a recurrent theme i n the Chase Variation . The point of the text is that with the k n i gh t still o n d5 neither of these queen moves offers anything. 7 Qf3?! can be met by 7 . . . Bb7 or indeed 7 ... c6, while 7 Qg4?! is well met by 7 . . . f5. The difference is that if 8 exf6 Black can recapture with the knight - i f he had already ex· changed on c3 and so had to use his queen to recapture he would not gain a tempo on White's queen (au contraire - White would shortly gain some time on his). Once . . . f7 -f5 is in much of the sting is taken out o f W h ite's Qg4 because Bg5 can now be easily met by . . . Be7 as th e g-pawn i s defended after Bxe7 Qxe 7.

7 cxb6 Of course the price Black h a s to pay for de l ayi n g . . . Nxc3 i s that W h ite can now exchange on d5. If he had done this on move 6 Black would have been able to equ alise easily with . . . d7-d6, but having committed himself to . . . b 7 -b6 Black will normally continue with ... Ba6 and a very sound game a fter 7 Nxd5 exd5 and either 8 Be3 or 8 cxb6 axb6 9 Nf3. The two themes Black has to remember (apart fro m the usual . . . Ba6) are to send his knight via a 6 and c 7 (after . . . c7 -c6) t o the superb e6square, and to meet f2 - f4 with . . .f7-f5. Usually it will suit him to trade his dark-squared bishop for a knight if he gets the chance, and he may either break with . . . d7-d6 or ... c 5 a t a suitable t i m e or continue with . . . b 5 and pressure o n the a - fil e .

7 . . .axb6 8 Nf3 N x c 3 9 b x c 3 B b 7 In fact I h a v e taken a liberty w i t h the move order a s the g a m e actu­ ally began with 6 . . . N xc 3 .

1 0 B d 3 d 6 1 1 0-0 Be7 1 2 Q e 2 N d 7 1 3 R d 1 White has also set u p with his rook on e l , and has played Bc2 instead of Qe2, but these nuances do not ch a n ge the fundamental nature of the position. Black's next re g ro u p i n g is one to remembe r .

13 . . . Ra5 (D iagram 1 0) 1 4 Bf4 Qa8 15 Ne l White doesn't care to have his pawns damaged

by . . . Bxf3 , but now Black has driven the knight away fro m its best position for an attack

on the kin g (w h ic h

1 60

is where White's chances must lie i n this varia-

T he C h a se V a ri a t i o n

t i o n , a n d a s a rule in t h i s structure the k n i g h t i s the most da ngerous piece in such an attack ) .

1 5 . . . 0-0 Note how Black has delayed . . . dxe5. Opening the d-file a n d freeing d 4 can only b e n e fi t White unless Black i s gettin g something in return.

16 Bc2 dxe5 Not 16 . . . Rxa2 17 Rxa2 Qxa 2 1 8 Bxh7+, of course, but now Black is ready for this.

17 dxe5 Nc5 18 Rd4 Rd8 1 9 h4 R a 3 20 Rad 1 Rxd4 21 Rxd4 Qe8 (Diagram 1 1 )

Diagram 1 1 T i m e t o regroup

{again)

Diagram 1 2 Closing out the enemy bishop

'!'here wasn't a concrete refutation o f 2 l .. .Rxc3 - or if there was I can't see it: 22 Bg5 Bxg5 23 Bxh7+? K.xh7 24 hxg5 can be m e t by 24 . . . Be4 shutting off the rook - but Kuczynski's choice i s very solid .

22 B g 5 Ra8 2 3 Qd2 B c 6 2 4 Bf6 A th em a tic blow; of course 24 . . . gxf6?? 25 Rg4+ mates, but Black re­ sponds well.

2 4 Bxf6 2 5 exf6 e5 26 Rg4 g6 27 h 5 e4 (Diagram 12) 2 8 Q £4 N e6 29 Qg3 •..

29 Qe3 Nc5 was a draw, but White can continue attacking.

2 9 ... Qd8 3 0 hxg6 hxg6 3 1 Q h 4 Of course 3 1 Bxe4 Bxe4 32 Rxe4 Qxf6 would allow Black's structural assets to come into their own.

3 I . . . Qd2 32 B b 3 B d 5 33 Bxd5 Qxd5 3 4 Rg3 Rxa2 3 5 R h 3 �-� It was not compulsory for Black to agree a draw: 35 . . . Qh5 36 Qxh5?! gxh5 3 7 Rxh5 Nf4 is decidedly better for h i m , while after 36 Qxe4 Qg5 (attacking f6) 37 Rf3 Qd2, for example, h e can certainly be confident about the future. Nonetheless a good ga m e in which attack and de-

1 61

S t arti n g O ut : A le kh i n e 's D e fe n ce

fe nce were equal to each other.

Statistics Pretty good. Black has scored 50%.

Theoretical? Not terribly - it is more important to know where to put the pieces af­ ter Nxd5 or in the game.

Theoreti cal Statu s The guru of 4 c5, Sveshnikov h i m se lf, re c e n tl y agreed a draw o n move 1 1 in the main line, which suggests Black is in good health . Person­ ally though I don't know that I care for the main line much. Black played well in the game a n d IF he can defend then White's split queenside pawns must tell i n an endga me; but i n other GM games in this line such players as Krasenkov and Morozevich have spent the whole game defending careful ly.

1 62

C h a pter E i g ht

O t h e r T r i e s fo r W h i te

• •

I nt rod uct i on I l l u st rative G a m es

S t a rt i n g O u t : A l e k h i n e 's D ef e n ce

I ntrod uction Finally we come to White's attempts to avoid the more well-known, conventional lines. Some of t h e possibilities featured below serve White better than others, w it h the more respectable earning their own illustrative games . I n G a m e 38 W h ite follows up 1 e 4 Nf6 2 e5 N d 5 3 d4 d6 with 4 f4 to bolster the centre i m m ediately, while G a m e s 39 & 40 see 3 Nc3 , Wh ite recapturing after 3 Nxc3 with the b - p a w n a n d d-pawn respectively . Game 4 1 investigates a kingside fi a nc h e tto treatment from White, n a m ely 3 g3, and Hector's interesting anti­ Alekhine 2 Nc3 is the subject of Game 42_ . . .

White has a good many other ideas, some of them fairly dangerous, but none of them quite warrants a ga me to itself.

A Random Bishop Move 1 e4 Nf6 2 e 5 Nd5 3 d 4 d 6 4 Bg5 (Diagram 1)

Diagram

Diagram

1

1

A club players' favourite

Hasn't that knight gone away?

This is without doubt the most historical of t he s e ideas, h avin g been p l ayed in the very first Alekhine's Defence game, Steiner·Alekhine, Bu da p e st 1 92 1 . As it was featured in a recent ChessBase m a g a z i n e it may be due for a revival , a l th o u gh I ra ther doubt it. Alekhine's own treatment can be improved upon today only slightly . ,

4 . . . h6 5 Bh4 dxe5 6 dxe5 Bf5 Alekh ine actually

played 6 Nc6 7 Bb5 Bf5 8 Nf3, reaching the same . . .

position, but White ought to h a v e played 8 Bxc6+.

7 Nf3 Nc6 8 Bb5 Ndb 4 ! 9 Na 3 Qxd l+ 1 0 Rxd1 Nxc2+ 1 1 Nxc2 Bxc2 In the game Alekhine had not inserted h7- h6, and White could h a ve obta i ned go od compensatio n with 1 1 Rd4 at this stage, pre v e nt i ng . . .

1 64

O t h e r T rie s f o r W h i t e

Alekhine's next move. With the bishop already on h 4 , however, the compensation is less a s i t cannot return to e3 to help with the attack. I n this position Black already stands well, for example:

12 R e i Probably 1 2 Bxc6+ fi r s t is a better drawing try .

1 2 ... Be4 1 3 N d 4 B x g 2 1 4 Rgl 0-0-0 An important resource.

15 Nxc6 Bxc6 e tc.

Anoth er Ra n dom Bishop Move 1 e4 Nf6 2 e 5 N d 5 3 d 4 d6 4 Bc4 (D iagram 2) This move, by contrast, is quite common in amateur chess. It can be met i n a number of ways a n d I present only the main l i ne here.

4 Nb6 • ••

Black can also develop with 4 ... c6 followed by . . . Bf5.

5 Bb 3 5 e6? had a brief period i n the sun in English club chess i n the late 1 970's but you are not likely to be lucky enough to encounter i t now. The refuta tion is 5 . . . Nxc4 6 exf7+ Kxf7 7 Qf3+ (White's sole point is that he can, contrary to appearances, win h i s piece back) 7 . . . Ke8 8 Qh5+ g6 9 Qb5+ Q d 7 1 0 Qxc4 Qg4, a n d White i s busted.

5 ... Bf5 The old main line 5 . . . dxe5 6 Qh5 e6 7 dxe5 is not so easy. From here Sax-Ghinda, Bath 1 9 7 3 continued thematically: 7 . . . a5 8 a4 Na6 (8 . . . Nc6 9 Nf3 Nd4 is sounder but still a little better for White; both 9 . . . Nd7 a n d 9 Bb4+ are alternatives) 9 Nf3 Nc5 1 0 Ba2 B d 7 1 1 Nc3 Nbxa4 12 Bg5 Be7 13 Nxa4 Nxa4 14 Rd1 Nc5 15 Be3 0-0 16 c3 Qe8 1 7 Bxc5 Bxc5 18 Ng5 h 6 1 9 Ne4 Be7 20 Nf6+ with a decisive attack since 20 ... g x f6 2 1 Rd3 mates. Ghinda's pawn grab i s extremely dangerous but has been tried by Baburin with su ccess. . . .

TIP: In these structures with a n open d-fil e and a pawn on e S facing o n e on e6, Nf6 + is White's most dang e ro u s attac k i n g idea. (Tai­ Gedeva n ishvil i , note to White's 4 th move, G a m e 3 9 , i s another ex­ ample). 6 Qf3

This is not forced, of course, but if White simply conti nues with 6 Nf3 e6 7 Q . Q Be7 he is in danger of eventually arriving a t an Exchange Variation with his bishop misplaced; alternatively he may find it blot­ ted out after . . . d6-d5.

6 Qc8 7 N e 2 ...

White's i m mediate ideas t o get to f7 do not work - 7 Bxf7+ Kxf7 8 g4 Nc6 9 c3 dxe5 , 7 g4 Be6 a n d 7 exd6 cxd6 8 Bxf7+ Kxf7 9 g4 Qxc2 are typical . After 7 N h 3 the sim plest is Bagirov's 7 . . . d5 (since 8 Bxd5

1 65

S t a rt i n g O u t: A l e k h in e 's D e fe n c e

Nxd5 9 Qxd5 Nc6 is grim for W h ite), after which Black can set up with . . . e 7 ·e6, . . . c7-c5, . . . Nc6 and so on, when White has a kind of French with his knights a n d k ing's bishop o n very much th e wrong squares, although 7 ... Nc6 - among others - is also possible .

7 ... e 6 8 Nbc3 This has done very well for W h ite, but to b e h on es t I don't understand why. 8 Qg3 dxe5 9 dxe5 c 5 (essential to free c 7 for the queen and al­ low . . . 0-0-0) 10 c 4 N c 6 led to a fas cinating game and a win for Black in Skytte-Baburin, Copenhagen 1 999. 8 . . . d x e5 Often Black has chosen 8 . . . d 5 here, which seems like a bad idea to me. White's knights are ideally placed for ha rassing the black bishop with Ng3, Ne2-f4, and if it h ides away o n h7 White can plant a knight on h 5 with a bind on the kings ide.

9 dxe5 c5!? This hasn't been played on my database. Instead Black has preferred 9 . . . N8d7 to be followed by . . . Nc5, which is also fi n e . But after 9 . . . c5 I really don't k n o w how White intends to meet the threat of 10 . . . c 4. The obvious attempt is 10 Ng3 Bg6 (not 1 0 ... c 4 1 1 Nxf5 exf5 1 2 Ba4+ Nxa4 13 Nxa4 Qc6 14 Qxf5 ! N d 7 1 5 Nc3 Qxg2 1 6 Rfl , which I think favours White, although l l . . .cxb3 1 2 Nd6+ Bxd6 1 3 exd6 bxc2 1 4 Qg3 is in­ teresting) but now I don't see what White intends to play with t h e s e ­ quence . . . c5-c4, Ba4+ Nxa4 , N x a 4 B x c 2 threatened. 1 1 h 4 h 5 changes nothing, defending the pawn with 1 1 Qe2 loses the g-pawn after 1 l . . .c4 1 2 Ba4+ Nxa4 13 Nxa4 Qc6, while 1 1 Na4 c4 12 Nxb6 axb6 1 3 Qc3 b5 is also grim , a n d al t o ge t h er I t h i n k Black stands very pleas­ antly.

Yet Another Random Bishop M ove 1 e4 Nf6 2 e 5 N d 5 3 d4 d 6 4 B e 2 (Diagram 3 )

Diagram

3

Better than it looks

1 66

D iagram

4

As bad as it looks

O t h e r T ri es for W h i t e

move is not fash ionable a n d never has been, but i t is quite dan­ gerous and, statistically, has scored very well. The idea is to keep open the option of f2-f4 and m e a n w h i l e prevent . . . Bg4. White's con­ crete intentions include 4 . . . g6 5 f 4 a n d 4 ... dxe5 5 dxe5 Nc6 6 N f3 Bg4 7 c 3 e6 8 Qa4 (or 8 0-0 Be7 9 Qa4), both of which are quite awkward to meet. Black's simplest reply is This

4 . . . dxe5 5 dxe5 Bf5 6 c3

6 Nf3 N b 4 amounts to a pawn sacrifice, though White does have some compensation.

6 ... c6 7 N £3 e6 8 0-0 Nd7 9 Nbd2 N7b6 10 Nd4 Bg6 1 1 N2f3 Be7 1 2 c4 Nc7 1 3 Be3 0-0 1 4 Qcl c5 etc. This line is certainly safe, but it is fair to say that it a i m s at equaliz­ ing gradually. In the stem game Romanishin- Bagirov, Jurmala 1987, it took the great Alekhinist until move 1 12 to persuade White that he h a d achieved this. I a m surprised we don't see more of 4 B e 2 .

A Really Random Pawn M ove 1 e4 Nf6 2 e 5 Nd5 3 c4 Nb6 4 a4 (Diagram 4) A bizarre idea, but presently e nj oy in g a vogue in certain circles . White's idea i s 4 . . . d6 5 a5 Nd7 6 e6!?. That's nothing to be afraid of, however. After, say, 6 . . .fxe 6 7 d4 Nc6 8 Nf3 g6 9 h4 Nf6, the po siti o n is similar to those i n G a m e 20, e xc e p t t h a t White h a s played a4-a5 a n d Black h a s transferred h i s knight from b 6 to f6, which m u s t surely favour Black. But the simplest answer is . . .

4 ... a 5 White c a n now head for Game 23 b y 5 d 4 d 6 6 exd6 cxd6 7 Nc3 g6, or alternatively he could play 6 f 4 with a Four Pawns Attack with the addition of a 2 - a 4 and . . . a 7 -a5, a l though I s uspect this favours Black, whose pi e c e s are more likely to use b4 than White's are to use b 5 . Likewise Black c ou ld and perhaps should meet 6 exd6 with 6 ... exd6. I n these sym metrical E x ch a n g e structures w he re Black still has the option of . . . Nc6 (as opposed to h avi ng played . . . c7-c6 already) White's a 2 - a 4 is pretty much in the na ture of a positional error (especially when Bl a c k still h as a k night on b8 ready to go to a 6 and b 4), since it means t h a t the typical sequence d4 -d5 Bx£3, B x f3 Ne5 followed by the exch a nge of da rk - s qu a red bishops on, for example, g5, will leave B l ac k with a n im m ovab l e k n igh t on c5 aiming forever at either th e a ­ pawn o r ( after b2-b 3) the b -pawn . This t u r n of events i s o f course not inevitable, but the need to avoid it restricts White's possibilities. I g ath e r, however, that aficionados of this line favour the strange

5 Ra 3 with the idea of taking the rook to g3 to inh ibit Black's kingside de­ I don't know of any games with this line but I'd suggest that Black proceeds with velopment.

5 . . . d 6 6 exd6 exd6

1 67

Sta rti n g O u t : Al e k h i n e 's D e fe n c e

After 6 . . . cxd6 I wonder if something with Re3 might justify White's play, for example 7 d4 g6 8 Nf3 Bg7 9 Be2 0-0 1 0 Bg5 Bg4 1 1 Re3 Nc6 1 2 d 5 . 1 f White really wants to play 7 Rg3 now then this can be met with 7 . . . Bf5 and 8 . . . B e 7 , when capturing t h e g·pawn just loses material to . . . Bg6 a n d . . . B£6. I don't see how W hit e is going to ma k e this work at a l l .

Random Tries after 2 e5 Apart from t h e main moves 3 c 4 a n d 3 d4, White h a s so m e others tha t are occasionally seen. A bizarre transposition to 1 b 3 occurs after 1 e4

Nf6 2 e5 N d 5 3 b3 g6 4 Bb2 Bg7 5 c4 Nb6, when Black will be fine as long as he notices that 6 Qf3 threatens c4-c5 a n d thus avoids 6 . . . 00?? 7 c5 resigns a n d plays instead, for example, 6 ... e6 followed by . . . Nc6, . . . 0-0, . . . d7-d6 a n d so on. TIP: W h enever B l a c k h a s answered c2-c4 wi t h N b6 and d o e s not control d5, h e needs to watch out for c4-c5.

It's easy to ignore c4-c5 because of the ghastly hole it creates on d5, but it can be strong. The clearest example of this is the dreaded Vo­ ronezh (c4-c5 is p a rticularly common after . . . e 7 - e 5 , dxe5 dxe5), but a diligent search will reveal others in the Four Pawns (5 . . . g5 6 N c 3 gxf4 7 Bxf4 Bg7 8 c 5 ! , or 5 . . . g6 6 Nc3 Bg7 7 Be3 0 - 0 8 c 5 ! , for example).

3 Nf3 is better but only really transpositional . White's idea is 3 . . . d6 4 Bc4, but 4 . . . Nb6 5 Bb3 Bf5 hardly leaves h i m a nything better t h a n 6 d4, which transposes to 3 d4 d6 4 Bc4 N b 6 5 Bb3 Bf5 6 d4, above. 3 Bc4 N b 6 4 Bb3 d6 is a nother transposition a s White cannot do better than 5 d4, leading to the same 4 Bc4 line. 3 Qf3 is occasionally played but really most moves will do for Black, for ex a m p l e Davies' 3 . . . c6 followed by . . . d7-d6 and either . . . g7-g6 and ... Bg7 or . . . Bf5 as appropriate.

White Avoids 2 e 5 have a couple of other second moves which are worth hav­ ing a l oo k at. The first is 2 d3. Alekhine u s ed to answer this with 2 ... e 5 . After 3 f4 B l a c k then plays 3 . . . Nc6. Th is is usually said to be nice for Black a fte r either 4 Nf3 d5 or 4 fxe5 Nxe5 5 Nf3 Nxf3+ 6 Qxf3 d5. Some off-piste complications a rise after 4 fxe 5 Nxe5 5 d 4 . Re­ treating must be bad, so presumably the game then goes 5 ... Nxe4 6 Qe2 d 5 7 dxe5 Q h 4+ 8 g3 Nxg3 9 Nf3 a n d now Black's simplest i s 9 . . .N x e 2 + 1 0 Nxh4 Nxc l . Black i s goin g t o lose t h e knight o n c 1 , o f course, b u t t h e weak e-pawn a nd th e two bishops ought to give h i m enough p l a y w i t h t h e t h r e e pawns a ga i ns t the piece. Failing that, however, White can obviously play some k i n d of reversed Phil idor or Pirc. White does

Black does of course have numerous options on move 2, another sound possibility being 2 ... d5. White can either try 3 e 5 Nd7 4 f4 with a rea-

1 68

Ot h e r T ri e s f o r Wh i t e

sonably promising anti -French set-up, o r 3 exd5 N x d 5 4 g 3 , which is likely to transpose to a Vienna line a fter 4 . . . e 5 5 Bg2 Nc6 6 Nc3, al­ though White has comm itted h i m self to d2-d3, which is not the case in the Vien n a . Alternatively, 2 c5 was played i n the classic posi­ tional crush Thomas-Alekhine, Baden-Baden 1 9 2 5 , which can be fou nd in Nimzovich's My System or i n Alekhine's own book of selected games. ...

The second is 2 Bc4, the idea of this being 2 ... Nxe4 3 Bxf7+ Kxf7 4 Qh5+ a n d Qd5+ winning the piece back. That is probably fine for Black, but alternatives include 2 . . . e6 followed by 3 d5, 2 d5 a n d 2 . . . e5 (transposing to the Bishop's Opening), all o f w h i c h should be equal, or the rather splendid 2 ... b 5 3 Bb3 (3 Bxb5 Nxe4) 3 ... c 5 4 e5 c4. ...

. . .

Il lustrative Games Game 38 D K u prei chi k • Z. Varga European Championship 1992

1 e4 Nf6 2 e 5 Nd5 3 d4 d 6 4 f4 (Diagram 5)

D iagram 5

D iag ram 6

White ignores the knight

A successful experiment

Rather a half-hearted looking syste m ('rubb ish' - Andrew Martin). If White wants to support his e-pawn in this way surely it was better to drive Black's knight out of the centre first, especially since Black h a s a specific answer which prevents this? Y o u might w e l l t h i n k s o , b u t so me strong players of a non-theoretical bent have played 4 f 4 regu­ larly (Victor Kupreichik i n particular has scored over 85% and chalked up such Alekhine specialists as Alburt, Varga and Kengis with it), so it deserves respect.

4 ...d x e 5 5 fxe5 c 5 6 N f3 White h a s n o decent alternative here. I t would hardly b e worth men-

1 69

S t a r t i n g O ut : A l e k h i n e ' s D ef e n c e

tioning 6 c4? Nb4 if so many players hadn't made th is mistake, i n ­ clud ing at least t w o 2 5 0 0 + grandmasters. The e n t i r e pu rpose o f Black exch anging on e 5 before goi n g . . . c7 -c5 was to prevent this by ensuring that now 7 a3 N4c6 8 d5? loses the e-pawn, while 7 d5 Bf5 8 Na3 e6 is also grim (9 d6 loses to 9 . . . Qh4+ 10 g3 Qe4+). TIP: W h e n White doesn't drive the d5 knight away immediately, a l ­ w a y s watch out for c2-c4 N b 4 i n stead o f

.••

Nb6

Another weak m o v e is 6 c3 because W h i t e doesn't have the time to maintain a pawn on d4, for example 6 c 3 Nc6 7 Nf3 Bg4 8 Be2 e6 a l ­ ready leaves Black th re ate n ing . . . Bxf3 a n d . . . Qh4+, winning t h e d ­ p a w n , s o the result will j u s t be a similar structure to t h e game but with White having played the rather useless c2-c3. 6 Bb5+ Bd7 7 Bxd7+ Qxd7 also fails to i m prove White's prospects.

6 ...c x d 4 7 Q x d 4 Nc6 8 Qe4 Kupreichik's latest twist. After 8 Bb5 Black should give the tempting 8 . . . Qa5+ 9 Nc3 Nxc3 a m i ss, a s 10 Bxc6+ bxc6 1 1 Bd2 i s deceptively difficult for h i m , a n d instead settle for 8 . . . Bf5 and ... e7-e6, a well­ respected recommendation of Kengis (which re c e nt ly gave Black an 1 8- m ove victory i n Pelikian-Milos, although it's fair to say that the gam e won't be in Pelikian's next book of selected games ) . In fact, this ins pires me to make one of my rare theoretical suggestions: it seems to me that Black might do well to consider 7 ... Bf5. 8 Bb5+ Nc6 then produces a familiar position, and White doesn't have any other par­ ticularly devastating manoeuvre available (8 Nc3 e6 doesn't seem dangerous, n o r does 8 c4 Qa5+ 9 Bd2 Nb4), so Black can obtain the set- up he w a n t s with th e bishop on f5 a n d ... e 7- e 6. The text is aimed at stopping this.

8 ... g6 Black could also have played 8 . . . e6. According to various people 9 c4 is then sligh tly better for White, although this doesn't seem like much to me after, for exa mple, 9 . . . Bb4+ 10 Bd2 f5 1 1 Qd3 ( 1 1 exf6 Nxf6 1 2 Qh4 Bxd2+ 1 3 Nxd2 N b 4 i s a poor idea) 1 l . . .N f4 1 2 Qxd8+ Kxd8 when Black can set up comfortably with . . . Ng6, . . . Ke7, . . . Bd 7, . . . Rhd8, . . . Rac8, . . . Be8 and so on. I would be more afraid of 9 a 3 .

9 B c 4 N b 6 1 0 B b 3 Bg7 1 1 0-0 B f5 1 2 Q £4 0 - 0 1 3 Nc3 (Dia gram 6) Both sides h a d alternatives, but these last few moves have been the most natural, and we've reached the sort of position White was aim­ ing for. Black has h i s chances, b u t White seems a little more comfort­ able to m e .

1 3 . . .Q b 8 A slightly unn atural move, b u t Black is short o f a good pl ace for the queen . Part o f Black's trouble is that h e hasn't managed to excha nge a pair of minor pieces and, as usual, this manifests itself in difficulty i n coordinating all fou r with the queen in Black's cramped quarters . Burgess s uggested 13 . . . N d 7 with the idea of sending the queen to a 5

1 70

Ot h e r T r i e s for W h i t e

instead, but I don't see that fundamentally affecting the position .

14 R e i N d 7 1 5 e 6 This is one trouble w i t h Black's s e t - u p . I t is nice to have t he queen's bishop active a n d the king's bishop pointing a t e5, but it's never pos­ sible to p l a y . . . e7-e6 for fear of losing the bishop to g2-g4, a n d without it this br eak is always i n the air.

15 .. . fxe 6 16 Qh4 Nc5 17 N g5 h6 18 N x e 6 B x e 6 1 9 Bxe6+ Nxe6 20 Rxe6 Evidently Black i s now worse. White is better developed and h as a structural edge.

2 0 ... Bd4+ 2 1 Be3 Rf6 22 Rxe 7!? (Diagram 7)

Diagram 7 Critica l . . .

Diagram

8

Black's king is in trouble

2 2 Qe4 would have maintained a slight adva ntage but Kupreichik is what would be called a hacker were he not a grandmaster.

22 ... R f4 22 . . . Nxe7 23 Bxd4 Rf7 24 Ne4 Nf5 25 Qg4 Nxd4 26 Qxg6+ Rg7 2 7 Nf6+ Kf8 28 Qh5 with compensation w a s t h e crucial line according to the bulletin . . . White c a n draw with 28 Nd7+ Kg8 29 Nf6+.

23 Qh3 Nxe7 24 Qe6+ Rf7 24 . Kf8 2 5 Bxd4 Rxd4 2 6 Qf6+. . .

25 Bxd4 Qc8 25 . . . Qf4 was essential , apparently.

26 Qe5 Kh 7 27 Ne4 Nc6 28 N f6+ Kg7 29 N e 8+ (D iagram 8) A cute way to drive the king into the open.

2 9. . . Kf8 3 0 Qh8+ Ke 7 3 1 Bc5+ Ke6 3 2 R e i+ Kd 5 3 3 Ba3 Qg4 34 Qg8 Qd4+ 35 Kh l Qf2 36 c4+ Kd4 36 ... Kxc4 37 Qxf7+ etc.

37 R d l+ K e 5 3 8 R d 5+ Ke6 39 Qxg6+ Rf6 4 0 Nc7 mate

171

Starting Out: Ale khine's Defe n c e

Stati s tics Dis t inc tl y encouraging on the face of it- Black has scored 5 1 % after 5 . c5. On the other h and part of this com es from White players play­ ing th e awful 6 c4?, whereas after 6 Nf3 White has a worrying 69%, although really this just emphasizes how little faith can be placed in these s m all samples . . .

Theoretical? Rare, an d not terribly anyway. In m y opinion m y s m all suggestion i s worth re membering, but I would say that.

Theoretical Statu s Not enough gam e s played to say - Black does need to take this sys­ tem s eriously, though. Game 39 0 Padevsky • M.Vu kovic Smederevska Palanka 1 971 1 e4 N£6 2 e5 Nd5 3 Nc3 (Diagram 9)

Diagram 9 White invites a tr ad e

Diagram 1 0 the centre

Pressure o n

Not a move which theory holds in great esteem, but in its time it has been preferred by greats such a s Mik hail Tal and Paul Kere s , so it must have something goin g for it. In fact, in my op i n io n it is just a s dangerous a s , s a y , t h e C h a s e Variation . 3 ... Nxc3

This is Black's usual move. White's two recaptures lead to radically different ty pes of game. We consider 4 dxc3 in Game 40. 4 bxc3

1 72

O t h e r T r i es f or W h ite

With this move White's intention is to build a big centre and use that space advantage to launch an attack o n the k ingside under the cover of his e 5 pawn. The game Tal-Gedevanishvili, Georgia 1 970, illus­ trates what White would like to happen better than a n y words of mine could, so I hope the reader will excuse m e if I quote it without much explan ation, especially as i t features a lesser-known Tal combi­ nation (reason enough to quote any ga me). This continued 4 . . . e 6 5 d4 d6 6 f4 c5 7 N f3 cxd4 (more cunnin g is 7 . . . Nc6, after which 8 Be2? cxd4 9 cxd4 dxe5 10 fxe 5 Bb4+ 1 1 Bd2 Nxd4 wins a pawn because 1 2 Bxb4 Nxf3+ and . . . Qh4+ recovers the piece; but of course Tal would h a v e avoided th is trap, for example with 8 Be3) 8 cxd4 dxe5 9 fxe 5 Bb4+ 10 B d 2 Q a 5 1 1 Bd3 N c 6 1 2 0 - 0 (when t h i s position a r o s e again in a game played much later White played the senseless 12 a3? here; after c ast l i ng White i s m uc h better) 1 2 . . . Bxd2+ 13 N x d 2 Qc3 ( 1 3. . . Nxd4? 1 4 Nc4 a n d Nd6+) 1 4 Rf4 0-0 1 5 Kh 1 Ne7 1 6 N e 4 Qb2 1 7 Rb 1 Qxa2 (rather a hot pawn!) 1 8 Nf6+ gxf6 1 9 Bxh 7 + Kh8 2 0 Rh4 Kg7 2 1 Qc 1 Ng8 22 Bxg8 1 -0. So h o w c a n Black avoid this sort of thing? One m e t h od is t o develop with 4 ... d5 and follow up with . . c7-c5, . . . Nc6, . . . Bg4 , . . . e7-e6, . . . Be7, keeping open options with the king, perhaps of . . . c5-c4 and . . 0-0-0, or even . . . Kd8-c7, or (if suitable) . . . 0-0. Assuming that White plays d2-d4 (he can also set up with f2-f4 and d2-d3) , then this looks like a twisted Winawer French ( 1 e4 e6 2 d4 d5 3 N c 3 B b 4 4 e 5 with a su bsequent . . . Bxc3+), save that Black has retained his dark-squared bishop i n ­ stead of a knight, a n d h a s developed h i s light-squared bishop outside the pawn chain. O n the face of it these look like improvements for Black, a n d indeed this line is supposed to b e fi n e for him, h as scored well over the years a n d h as been the choice of various Alekhine ex­ perts a n d grandmasters . If you like French-type pos i t i on s this is a sensible way to play. However, Black has to be aware of two differ­ ences fro m the Winawer. First, the kni ght on e 7 in the Winawer i s a u seful piece for two reasons. On e, it holds u p a n f4 -f5 advance (so much so that f2 - f4 i s almost never the right plan for White in the Winawer). Two, by threatening to go to f5 i t puts pressure o n White's centre, and in particular means that White's dark-squared bishop cannot easily come to e3 to bolster the centre. A knight on e7 would i n fact have more active possibilities than the bishop. Secondly, a s any French p lay er will c o n fi rm if Black exchanges his 'bad' light-squared bishop for a knight, he is in danger of finding that White's light­ squared bishop scythes him down once the ga m e opens u p . So the bishop being outside the pawn chain i s not an unmixed blessing (it also prevents any restraint of f5 and g4 by . h7-h5 or . . . g 7-g6: w h e n White plays h 2 - h 3 , probably Black w i l l h a v e t o p l a y . . . B h 5 . White may then play g2-g4, which will further support an f4-f5 break. These reflections suggest that White o u gh t t o play f2-f4 , N f3 , Be2 (not Bd3 when Black has less pressure on the centre than h e does in the Winawer he is very likely to want to play . . . c5-c4 , when Bd3 will sim­ ply have lost a tempo), Be3, 0-0, h2-h3, g2-g4 a n d then open h i s eyes .

.

,

,

. .

1 73

Sta rt i n g O u t: Al e k h i n e 's D e fe n c e

a n d s e e w h e t h e r he is t hr ea te n i n g f4 - f5. In Y u d a s i n - E h l vest, Biel 1 99 3 , p l a y d i d indeed proceed 4 bxc3 c 5 5 f4 N c 6 6 d 4 d 5 7 Nf3 Bg4

8

B e 2 e6 9 0-0 B e 7 10 h3 B h 5 1 1 R b l Q c 7 1 2 Be3 c4 1 3 g4 Bg6 1 4 N e 1 , a n d White w a s i n d e e d t h r e a t e n i n g t h e highly u n p l e a s a n t f4 - f5. Blac k, a 265 0 GM, went o n to be c r u s h ed . Of c o u rse one g a m e proves n o t h i ng a n d various a lt e r n a t i ves ( l l . . . Qd 7 - D a v i e s ; 14 . . . Be4 or 14 .. . f5 - B u r ­ g e s s ) have been p r oposed, but I t h i n k t h e sy s t e m i n t h e m a i n g a m e i s a b i t e a s i e r to p l a y . I n fac t i f I w a s g o i n g to play . . . d 5 I t h i n k I m i g h t be i n c li n e d to d e v e l o p t h e bishop i n s i d e t h e p a w n c h a i n . I f White i s g o i n g to p l a y for f4 -f5 t h e n i t w i ll come i n t o i t s o w n a ft e r t ha t , a n d i t a l s o h a s l o n g - t e r m prospects e i t he r o n a 4 or o n g 6 a fter . . . f7 - f6.

4 . . . d6 5 f4 Plainly t h e o n ly move. White c a n n o t c on t e m p l a t e 5 exd6 cxd6, expos­ i n g h i s c - p a w n s a lo n g a n o p e n fi l e , nor 5 d 4 d x e 5 , exchang i ng queens a n d leaving White n o c om p e n s a t i o n for h i s w e a k nesses, a n d 5 Nf3 a l s o l e a d s to the exchange of the e 5 - p a w n a n d t h e loss of White's space a d va n t a g e .

5

...

g6

T h i s i s t h e difference fro m t h e T a l g a m e - B l a c k b l u nts i n a d v a n c e t h e b i s ho p he a n t ic ipates a pp e a r i n g o n d 3 , a nd d e velops h i s o w n so as to pressurise t h e centre with . . . d 7 -d6 a n d . . . c 7 - c 5 . T h e g a m e will r e ­ s e m b l e a strange k i n d of G r u n f eld , e x c e p t t h a t Blac k h a s a n extra p a w n o n d 6 and White o n e o n c2.

6 Nf3 Bg7 7 d4 0-0 8 Bd3 8 Be2 would be feeble - Black c a n d e velop m u c h a s i n t h e g a m e b u t p l a y . . . Bf5 i n s t e a d of . . . Bg4; t h e b i s h o p i s v e r y s t rong h e r e , c o n s t a n t l y t h r e a t e n i n g c 2 a f ter a n e x c h a n g e o n d 4 .

8 Bc4 i s also p o s s i b l e a n d r e ­

quires a l i t t l e c a r e . Black s h o u l d n o t b l i t h e ly p l a y . . . c 7 - c 5 x d 4 and a l ­ l o w t h e bishop to s e t t l e o n b 3 . I t i s d a n ge r o us there w i t h t h e f-file open, and i f Black removes i t with . . . Na5xb3 the recapture with t h e a p a w n w i l l restore the i n t egrity of White's p a w ns a n d g a i n White fur ­ t h e r t i m e w i t h t h e a t t a c k o n t h e a - pa w n w h i c h W h i t e c a n use to d e ­ fe n d h i s c e n t r e . I n s t e a d I l i ke B l a c k ' s p l a y i n t h e o l d g a m e C a n a l ­ C o l le, M e r a n 1 92 6 , w h i c h w e n t

8 . . . c5 9 0 - 0 Q c 7 1 0 Q e 2 Nc6 w h e n

Blac k w a s a l r e a d y sta r t i ng to annoy t h e b i s h o p a n d t h i n k i ng of . . . Bf5 and . . . N a 5 to force its exchange on d3 for the q u e e n ' s bishop (a highly desirable exchange fo r Blac k ) , so much so that C a n a l lost quickly af­ t e r l a s h i n g out w i t h 1 1 e6 f5 ! (sorely rest r a i n i ng W h i t e 's queen's b i s h o p a n d k i n g 's rook - note h o w usefu l it is t ha t Black h a s not e x ­ c h a n g ed p a w n s a n d t h us t h e r e i s s t i l l a pa wn o n f4) .

8 . . . c5 Actually I t h i n k it is more accurate for Black to p l a y

8 . . . d x e 5 . Exactly

n o w , once White's b i s ho p i s commit ted to d 3 (one wouldn't want it c o m i ng to c4 after opening the f-file w i t h . . . d xe5) and b e fo r e Black commits h i mself to . . . c 7 -c5. I t i s useful to know whether White i s go­ i n g to recapture w i t h t h e 'd' or t h e f- pa wn. In the for m e r case . . . c7 - c 5

1 74

Ot h e r T r i e s f o r W hi t e

is not necessarily s om ethi n g Black wants to play. 8 . .. dxe5 9 fxe5 will transpose to t h e game, but after 8 . . . dxe5 9 dxe5 I believe 9 . . . Nc6 10 0 - 0 f6 is a very good idea - it hasn't been played but by openi ng the long diagonal u p i n this way it seems to me that Black exposes the weakness of the c-pawns. In such a p l an . . . c 7-c5 is a complete waste of a te m po. I won't fill up the page with variations, but I think Black is already at least equal h ere , e.g. 1 1 Bc4+ Kh8 12 e6? Na5 13 Qe2 Qd6, threatening both ... Qc5+ and ... Nxc4, is good for Black, and 1 1 exf6 Bxf6 or 1 1 Be4 Bg4 are also fi n e .

9 0-0 d x e 5 1 0 fxe5 Keres preferred 1 0 dxe 5 , a now forgotten but dangerous variation. White's c - pawns then do not make a pretty imp ression, but they are hard to attack, and they also help by ensuring that h i s minor pieces will have at least one stable post o n the d - file to keep out Black's rooks. Meanwhile, once h e has developed his bishop to e3, his queen to e 1 o r e2 and h i s queen' s rook to dl, White has active ideas such a s Ng5 -e4, Bf2 - h4 o r Q h 4 t o consider, while for Black it i s harder t o fi n d a constructive p l a n .

1 0 . . .N c 6 1 1 B e 3 B g 4 (Dia gram 1 0) This position has been reached in many games a nd is fi n e for Black. Right now d4 is en prise - hence White's next, to meet 12 . . . cxd4 with 13 Bxc6.

1 2 Be4 Qa5 13 Q e 1 Possibly 1 3 Qd3 i s better, mee tin g 13 . . . Rad8 with 14 Q c 4 . This oc­ curred in a game where Solozhenkin was Black, and he lost badly, but why he rejected the obvious 14 . . . cxd4 15 cxd4 Rc8 at this point I don't know. Black seem s to be reasonably placed then.

1 3 ... Rad8 1 4 R d 1 B x f3 1 5 gxf3 15 Rxf3 loses material to 15 . . . cxd4 16 cxd4 Qxe l+, decoying t h e rook fro m the defence of d4.

1 5 .. . Qxa2 16 f4 e 6 17 R a 1 Q c 4 18 Bd 3 Q d 5 19 dxc5 f6 An i mportant theme. T IP : I n t h i s l i n e i t I s u s u a l l y essential t o b r e a k W h i t e ' s centre, a n d t h a t m e a n s e m p l oying every p a w n b r e a k a va i la b l e , e v e n i f i t d o e s weaken e 6 .

Otherwise White can build up in peace with his space advantage while the g7-bishop is inactive, as can also happen in the Grunfeld.

20 exf6 Bxf6 21 R d 1 Be7 22 Q f2 Qh5 White's exposed ki n g a n d ragged pawns give Black a near decisive advantage.

2 3 Rb 1 e5! 24 Be2 24 Rxb7 exf4 wrecks White. 24 . . . Qf5 25 R x b 7 exf4 2 6 Bd4 Bf6 2 7 Rdl f3 28 Bc4+ Kh 8 29 Bxf6+ Qxf6 3 0 Rxd8 Rxd8 31 Bfl Rf8 3 2 R d 7 Ne5 3 3 Rd6 Q g5+

1 75

S t a r t i n g O ut : A le k hi n e 's D e fe n c e

34 Kh 1 R£4 35 c6 Rf7 36 Re6 Re7 37 Rxe7 Qxe 7 38 Qd4 h5 3 9 B d 3 Kh7 4 0 Q e 3 a 5 4 1 Bf1 Qc7 4 2 h 3 a 4 4 3 Qc5 Nxc6 44 c 4 Qd 7 45 c 3 Qe6 4 6 Kg1 a3 4 7 Qxa3 Q e 3+ 48 Kh 1 Qf2 0- 1

Stati sti cs 4 bxc3 as a whole has brought White 52% (4 . . . d6: 54% White, 4 ... d5: 4 7% White).

Theoretical? Not really. Black's moves (attacking d4) i n t h e main l i n e flow very ob­ viously. Probably the only things worth comm itting to m emory are how to meet 8 Bc4 and the timing of my idea 8 . . . dxe5 ( a s su m i n g B lack likes it! ) , but these can be worked out fro m the ideas of the l i n e .

Theoretica l Status Comfortable. S t ron g Black players s e e m to prefer 4 . . . d 5 , but I don't really understand why. Black has always done fine with 4 . . . d6. Per­ haps it is a question of taste. Game 40 0 Ada ms • S . A gde stei n Oslo 1994

This gam e is annotated by Igor Stohl in his ma rvellous Instructive Chess Masterpieces, and my notes are based on h i s . I cannot recom­ mend this book too highly.

1 e4 N f6 2 e 5 Nd5 3 Nc3 Nxc3 4 dxc3 With this move White annou nces a very different strategy - essen­ tially he aims to develop his pieces as fast a s possible a nd derive some benefit fro m piece play. He recognizes that the e- pawn cannot be maintained after 4 d6, but hopes to gain further time while Black is exchanging it. . . .

4 ... d5 Just as in the last ga me, Black can also pl ay 4 . . . d6. A couple of rea ­ sons not to, though, are first that it tends to lead to a rather dull game (if Black is lucky) and, secondly, that this is the easiest varia­ tion i n which Black can lose i n 2 0 moves in the whole of Alekhine's Defence. Indeed after 5 Bc4 dxe5?? 6 Bxf7+ he can manage i t in six. WAR N I NG : In any Bc4 1 in e , always check for Bxf7+ before p l a y i n g e ither . . . N b6 or . . . dxes.

However, after 5 Bc4 Nc6 6 Nf3 dxe5 7 Qxd8+ Nxd8 8 Nxe5 f6 a n d . . . e7-e5, or 5 Bf4 Nc6 6 Nf3 dxe5 7 Qxd8+ Nxd8 8 Bxe5 c 6 (again fol ­ lowed by . . .f7-f6, . . . e7-e5, . . . Bg4 , . . . Be7, . Ne6 and so o n ) Black should be fine. The slightly dangerous line i s 5 Nf3. Now Black can either go in for the poorly charted 5 . . . Nc6 6 Bb5 Bg4 (6 . . . Bd7 7 Qe2 is more dif­ ficult but quite possibly also adequ ate) or play the conventiona l ..

1 76

O t h e r T r i es for W h ite

'equaliser' 5 . . . dxe5 6 Qxd8+ Kxd8 7 Nxe5 Ke8. Black has to be a bit carefu l here. With h is king where it is he i s probably well advised to settle for ... N d 7 , . . . e7-e6, . . . Bd6 type plans rather than aim for ... f7-f6 and . . . e7-e5. Meanwhile White may try the irritating 8 Bb5+. Because 8 . . . c6 weakens d6 and hence prevents the above pla n , and 8 . . . Nd7 leaves Black a bit tied up after 9 Bf4, Black has replied 8 . . . Bd7, but th e n 9 Nxd7 Nxd7 gains the two bishops and leaves Black with a slightly thankless position, a l th o u gh those who like endga m e grinds might find it appealing.

5 Q f3 (Diagram 1 1 )

Diagr a m 1 1 White prepares a quick 0-0-0

Diagram 1 2 Black clamps

down on

c4

White has m a ny other ways to develop, of course, for example 5 c4 c6, 5 Bd3 c5, o r 5 f4 Bf5, but in a n y event Black's plan of ... c 7-c6, ... Bf5/g4, ... e7-e6, . . . Nd7 shou ld give him a decent position. The text attempts to stop this plan by p re v e n t i n g ... Bf5.

5 ... c6 Davies reco m m ended 5 . . . c5 with enthusia s m . T h e trouble may be that Stohl's 6 Be3 more or less obliges Black to play 6 . . . e6, undesirably blocking in the queen's bishop. Agdestein prefers to keep open the possibility of d ev e l o p i n g this o n f5.

6 Bf4 I t is a pity to allow the bishop to f5, which rather negates Whi te's id e a with 5 Qf3, but White is a little short of a convenient developing move, especially since 6 Bd3 allows 6 . . . Nd7, intending . . . Nc5, w h e n 7 e6 Ne5 8 exf7+ Nxf7 9 c4 e5 1 0 c x d 5 cxd5 i s given a s okay for Black b y Adams, 1 1 B b 5 + Bd7 1 2 Qxd5?? losing to 1 2 . . . Qa5+. Nonetheless I can't believe 6 Bf4 is the best move. I n fact I am t e m pt e d to say th a t even a 'pass' would h ave been better - when you look at the measu res Adams later has to take to try and blot out Black's queen's bishop you may agree.

1 77

S t a rt i n g O u t : A l e k h i n e 's D e f e n c e

6. . . B f5 7 0-0-0 e6 8 Qg3 h6 9 h4 Qa5 1 0 K b 1 b 5 ! ( D i agram 1 2) It may not be my place to say so, but Adams h a s not played the open­ ing too effectively and I suspect Black i s already better. The text rules out c3-c4, and without that move it's not easy to see how White is go­ ing to counteract Black's m a ting attack on the queenside.

1 1 N f3 c5 The idea of Davies' 5 . . . c5 is in effect to get this k i n d of position a tempo ahead of this game, but even as it is Black is doing very well.

12 h 5 Na6 13 Nh4 Bh7 Apparently Agdestein pointed out 1 3 . . . Be4 after t h e game, the idea being that if 1 4 f3 Bh7 White's rook cannot be effectively deployed on h3. This looks like a very difficult idea to meet indeed.

14 Q g4 The start of a desperate plan.

1 4 ... b4 15 Bd2 c4 1 6 Ng6! (D iag ram 1 3)

Diagram

13

Active d efence

Diagram

14

The t id e has turned

This was White's idea. The razor bishop on h7 simply has to be blunted.

16 ... Rb 8 Of course not 1 6 . . . fxg6? 1 7 Qxe6+ etc. Black has various ways to try and take the k night, but any time the sequence . . . fxg6, hxg6 Bg8 oc­ curs White must have at least reasonable compensation as Black has a couple of pieces totally out of the game, possibly for ever.

17 R h 3 N o t 1 7 Nxh8 b3, but . . .

1 7 . . . Be7? Again Agdestein pointed o u t 1 7 ... Q b 6 directly after t h e game. It was thought that this won for Black; the variations are too nu merous and

178

Ot h e r T rie s f o r W h it e

complicated to give here, but White's basic problem is that h i s knight i s now e n prise as t h e queen defends e6, and that 18 Nxh8 still goes down in flames after 18 . . . b3 19 Kal bxc2 20 Bel Qxb2+ 2 1 Bxb2 Ba3 22 Bxa3 Rb l + and mates. 18 Be3 Bc5 is not so good either. 18 Be l fxg6 1 9 hxg6 Bg8 seems t o be critical, though. Instead Stohl spends a couple of col­ umns showing that 1 7 . . . b3! gave Black very good winning chances.

18 Bxc4! d xc4 1 9 Qxc4 Qc5 Black has already m issed the boat, but 19 . . . Bxg6 was a better chance according to Stohl.

20 Qxc5 Nxc5 21 N x h 8 Ne4 22 Be l bxc3 23 b3 Bb4? This loses. According to Adam s 2 3 . . . Bf5 24 Rf3 Bg4 25 Nxf7 Bxh5 26

Nd6+ Nxd6 27 exd6 Bxf3 28 gxf3 Bf6 29 d7+ Ke 7 was about equal, but Stohl gives the unnatural 2 4 Rhl as all but wi nning.

24 R f3 (Diagram 14) Decisive. Black cannot defend the f-pawn a s 2 4 . . . Bf5 2 5 N x f7! comes anyway. 24 . . . Nd2+ 25 Bxd 2 RdS

2 5 . . . cxd2 26 Kb2 a n d 2 7 c3 changes nothing.

26 Nx£7 R x d 2 2 7 Rei Ba3 28 N d 6+ Ke 7 29 Rxc3 Bxc l 30 Kxc l Rx£2 3 1 g4 R g 2 3 2 R c 4 R e 2 3 3 R c 5 R g 2 3 4 N b 5 Rxg4 3 5 Rc7+ Kd 8 3 6 Rxa7 R g 5 37 Nd4 Kc 8 3 8 Nxe6 Rxh5 3 9 Rxg7 B e 4 40 Rg8+ 1 -0 Adams' greater p ra c tic al strength decided this game, but White needs a fundamental i m provement before he treads this path again .

Stati stics Overall White h a s scored 55% after 4 dxc3, although h e makes 57% after 4 . . . d 5 and only 53% a fter 4 . . . d6.

Theoretica l ? Not a t all - certainly not after 4 . . . d5. 4 . . . d 6 needs a bit more study.

Theoretica l Status H a s b e e n considered equal for a very long time, b u t this variation (af­ ter . . . d7-d6 a t any rate) will suit someone with White who likes

queenless semi-endi ngs. For example the Bulgarian GM I van Radulov scored 5 . 5/6 with it. For Black the choice i s definitely a question of taste - 4 . . . d5 is a more fighting move. Game 4 1 0 Roze ntal i s • Adams Bundesliga 1998

1 e4 N f6 2 e5 N d 5 3 g3 (Diagram 1 5) This is the only other serious try for White on move 3. At first sight a

1 79

S t a rt i ng O u t: Al e kh i n e ' s D e f e n c e

k ingside fianchetto doesn't seem to go with e4-e5, a n d the same is also true at second and third sights. White does, however, have a cun nin g idea, which is that the natural (and common) sequence 3 . . . d6 4 exd6 cxd6 5 Bg2 Nf6 6 d4 g6 7 Ne2 Bg7 8 0-0 0-0 9 c4 Nc6 10 Nc3 produces a position variously described as a n English or a Fianchetto King's Indian, but either way supposed to be slightly disagreeable for Black.

Diagram

15

An un expected fianchetto

Diagram

16

A barrier

3 ... d 6 4 exd6 An important moment. Any of the three recaptures is reasonable. Black can perfectly well adopt the line given above with 4 . . . cxd6. He should then continue with 10 . . . a6, . . . Bd7, . . . Rb8 and aim for . . . b7-b5. Alternatively 4 ... Qxd6 is solid, for example 5 Bg2 e 5 6 N b c 3 Nxc3 7 bxc3 gives Black an extra tempo ( . . . Qd6) on the 3 g3 Vienna. That isn't a n u n mitigated bonus, since with the queen on d8 his normal move would be . . . Bd6, but Black ought to be fine a fter, for example, Baburin's 7 . . . c6, or 7 . . . Nc6. White has generally done best with a plan involving Ne2, 0-0 a n d Rb l , a voiding any adventures with h2-h3 and g3-g4 in reply to ... Bg4 .

4 . . . exd6 5 Bg2 Nf6 6 d 4 Be7 6 . . . d5 a n d 7 ... Bd6 is also possible. 7 Nf3 Rozentalis has also played 7 c4 against Miles, after which Miles' 7 . . . d5 was more than adequate - White has either to accept the isolated d­ p a w n after 8 cxd5 (which seldom goes w i t h a fianchetto) or - as Rozentalis did - play 8 c5, when the game continued 8 ... 0-0 9 Ne2 b6 10 cxb6 axb6 1 1 0-0 Ne4 1 2 f3 Nd6 (a fine regrouping) 13 Nbc3 c6 1 4 Bf4 ReS 1 5 b 3 a n d Black was already a t least equal and could, per­ haps, even have posed White problems with 1 5 . . . Nf5.

7 0-0 8 0-0 d5 9 Ne5 ...

1 80

Oth er T ri e s f or W h i te

Not the only move, o f course, b u t i f White wants t o aim for anything h e surely has to try this.

9 . . . c6 (Diagram 1 6) TI P : It i s often use ful to e rect a sturdy pawn barrier s u c h as this a g a i n st a fianchettoed bishop.

In the present case this s trategy is especially effective, since if White wants to attack the p a w n barrier he has to play c2 -c4, which involves weak e n i n g h i s d-pawn.

1 0 Nc3 Ndb7 1 1 f4 Nxe5 1 2 d xe5 There must have been considerable psychology involved here. Four years before Rozentalis had reached this same position against Appel a nd h a d continued with 1 2 fxe5 Ne8 1 3 Qh5. Comme ntators had now suggested 1 3 .. .f6 14 Bf4 as Black's best, still with a slight a dvantage for White. It is fair to assume Adams didn't agree with this. Adams seldom plays the Alekhine a n d Rozentalis i s very loyal to this system. Adams surely k n e w this a n d was a i m i n g fo r t h i s position; Rozentalis must have known this ; Adams must have known that he knew, and so on. In fact Ada ms had already twice played the Alekhine agai nst Rozentalis, and on both occasi o n s Rozentalis had ducked this l i n e and pr e ferre d 2 Nc3 e5, so both players must have been ready for this po­ sition. M ea nwh i l e, after 12 fxe5 Ne8 13 Qh5 f6 14 Bf4 Qb6 I can't really see how White pr oposes to demonstrate compensation for the pawn he's about to lose. Judging from this game 12 dxe5 isn't much good either, so possibly the whole Ne5 manoeuvre is m isguided and White would do better to settle for equality earlier. T h e fine finish of Rozentalis-Appel, i ncidentally, is wonderfully a nnotated in Jonathan Rows on ' s The Seven Deadly Chess S ins ( 2 5 a4!).

12 ... N g4 ! (Diagram 1 7)

Diagram

17

The aggressive option

D iagram

18

White's d efences collapse

Far be tter than 12 . . . Ne8. Black can take the initiative on the king-

1 81

S t a rt i n g O u t: A l e k h i n e 's D e fe n c e

side. 1 3 . . . Bc5+ is threatened.

13 Kh1 f6 14 h3 Nh6 15 exf6 Bxf6 1 6 g4 Rozentalis surely must have thought that he was forcin g Black back, but in fact the contest is already over. Evidently White should have pl ayed more conservatively during the last few moves, but it i s clear that h e was already struggling to equalise.

1 6 ... Bxc3 17 bxc3 Bxg4 ! (D iagram 1 8 ) It's h a r d t o imagine Rozentalis h a d n ' t seen t h i s a t a l l , b u t h e must have miscalculated a few moves back. This is really quite a simple sacrifice for players at this level to calculate.

1 8 hxg4 Qh4+ 1 9 Kgl Nxg4 2 0 R£3 The only try as 20 R e l Qf2+ 2 1 K h 1 Rf6 2 2 Qxg4 Qxe l + 2 3 K h 2 Rh6+ 24 Bh3 Re8 is crushing.

20 ... Rae8 2 1 Bd2 White doesn't have any defence to the slow p l a n of doubling o n the e­ fil e and invading on e 2 .

2 1. . . Re4 2 2 f5 Rfe8 2 3 f6 gxf6 2 4 B h 3 R e 2 0 - 1 B l a c k threatens 2 5 ... N h 2 ( a m o n g others) a n d t h e r e i s n o t h i n g t o b e done.

Sta t i st i cs White scores 54% with 3 g3 as a whole but has usually been higher rated. After 4 . . . exd6 he scores 59%, after 4 . . . cxd6 it is 54% a n d a fter 4 . . . Q x d 6 d o w n to 4 9 % (completing a hat trick in this chapter where a l l my recomme ndations have scored worse than the alternatives fo r Black!).

Theo retica l ? N o - one contest has seldom followed a previous one.

The o ret ica l Status Equal - the better player will win. The main ga m e with 12 fxe5 i s supposed t o be better for White, but we can assume Adams a t least didn't believe this, a n d probably Rozentalis as well. For what it's worth , I don't either. Game 42 0 He cto r • A nag nosto u Komotini 1 992

1 e4 Nf6 2 Nc3 (Diagram 1 9 ) What k i n d o f move i s that?? White ducks the theoretical debate a n d allows Black the option o f 2 . . . e5, transposing t o the Vienna o r (after 3 Nf3 Nc6) the Four Knights, both of which are supposed to be equal for Black and have been for quite so me decades.

1 82

Ot h e r Tri e s f o r W h ite

Diagram

Diagr am

19

White 'ignores' the knight

An

20

unusual position

2 Nc3 is actually extremely common below international level - I have met it i n over 50% of m y games with the Alekhine - so it is worth preparing for and, personally, I recommend 2 . . . e 5 . I believe this is the strongest o p t io n and, a s luck would have it Every m a n p ublishes two excellent books on the Vienna and Four Knights b y Gary Lan e and ,

Jan Pinski respectively . . .

It m a y seem l i k e a nuisance t o have t o learn a defence to t w o other whole openings when you want to play the Alekhine, but after all if you were going to play l . e5 you'd need a defence to these openings as well as the Scotch , Ponziani, King's Gambit, Ruy Lopez, G uioco Piano, Danish Gambit and so on. Moreover a t least some of those 2 Nc3 players are b l u ffi ng and know nothing about either of these openings . .

themselves. For exa mple Chris Baker's Startling Opening Repertoire recom mended 2 Nc3 and airily assured readers that '2 . e5 is u ncom­ mon'! In any event there's a lot to b e said for tackling the opening by playing the moves you believe to be the best. .

.

The only independent move, though, is 2 . . . d5, and that is the su bject of this gam e .

2 d5 3 e5 . . .

White c a n also play 3 exd5 Nxd5 4 Bc4 (or 4 g 3 b6!?, or 4 d4/Nf3 Nxc3 5 bxc3 g6, or 4 Qf3 Nb6/Nb4). 4 Bc4 needs a modicum of care and at­ tention - in particular 4 . . . e6? 5 Bxd5 exd5 6 Qe2+ winning a pawn (6 . . . Be6 7 Qb 5+ ; 6 . . . Be7 7 Qe5; 6 . Qe7 7 Nxd5) should be avoided (al­ though Black does have some compensation in the second line). ..

WARN I NG : This i s a very popular tra p

The safest is 4 . . . Nb6 5 Bb3 Nc6 (and certainly not 5 . . . e 5 6 d 3 N c 6 7 Nf3, when defending f7 is awkward, especially since 7 . Bg4 8 h3 B h 5 9 N xe5 is either the e n d o r t h e start of a very unpleasant experience involving having your king cha sed down to e2 or so after 9 . Bxd l 1 0 . .

.

.

183

S t a rt i n g O u t: Al e k h i n e 's D e fe n c e

Bxf7+ Ke7 1 1 Bg5+ Kd6 12 Ne4+ Kxe5 1 3 f4+ Kd4 1 4 Rxd 1 a nd so on, depending on how yo u look at it) . Mter 4 Bc4 N b 6 5 Bb3 Nc6 6 N£3 Black can develop either with ... Bf5 and . . . e 7 -e6 or with ... g7 -g6, play­ ing ... e 7-e6 only if forced by Ng5. In either case he has n o p a rticular probl e m s .

3 . .. N e 4 Black h a s two alternatives. 3 . . . d4 4 exf6 d x c 3 5 fxg 7 cxd2+ allows e i ­ ther the dull

6 Qxd2 Qxd2+ 7 B x d 2 Bxg7, w h e n White continues with

0 · 0-0, N e2 and Bc3 and can claim a slight advantage based o n his

or the d a nge ro u s 6 Bxd2 Bxg7 7 Qf3, a favour­ Swedish G M Jonny Hector, when 7 . . . Bxb2 is rather rash, and otherwise W h ite continues with 0-0-0 and i s decid­ e dl y better develope d . 3 . . . Nfd7 may well be best. The d-pawn is en prise, of course, but 4 Nx d5 Nxe5 5 Ne3 c 5 6 b3 Nec6, followed by ... e7-e5, is fine for Black. Alternatively 4 e6 fxe 6 5 d4 c5 6 Nf3 Nc6 7 Bb5 g 6 8 dxc5 Bg7 9 Be3 Qa5 1 0 0-0 is probably about equal. 4 d4 c 5 invites either this line or a t a m e line of t h e French after 5 Nf3 e6, while 5 Nxd5?! cxd4 6 Qxd4? N b 6 sees White lose a piece. However, n ote that after 3 . . . d4 4 Nce2 Black probably has no better than 4 . . . N e 4 returning t o the ga me. more com pact pawns, ite of the i n g e n i o u s

4 Nce2 Nxe4 dxe4 5 d 4 allows either equality v i a the tedious 5 . . . exd3 or the more s p ecu l a tive a n d vi rtu a l ly uncharted 5 . . . Bf5, w h il e 4 Qf3 Nxc3 5 dxc3 is G am e 40. The text is the critical m o v e , aiming either to trap the knight or to drive it somewhere hopeless w it h gain of tempo. 4

4 ... d4 4 . . . f6 is a n alternative, but I don't think Black is doing very well afte r 5 d3 Ng5 6 Bxg5 hxg5 7 h4 g4 (7 . . . gxh4 8 Nf4 g6 9 Rxh4 has been looke d at askance since a game Vorotnikov-Kengis, Riga 1 983, which is worth p layi n g t h ro ugh if you can find it on a database). White can play 8 Nf4, w he n his plan is simply to take the g-pawn with Be2. Black has v a ri ou s ideas - 8 . . . g6 9 d4 (preparing a retreat for t h e knight) 9 . . . Bh6 1 0 N d 3 N a 6 1 1 B e 2 c5 12 Bxg4 a n d 8 . . . c5 9 B e 2 N c 6 · 10 Bx g4 Q c 7 (not 1 0 ... Nxe5 1 1 Bxc8 a n d 1 2 Qh5+) 1 1 Bxc8 Qxe5+ 1 2 Q e 2 Qxf4 1 3 N h 3 are a couple of ways of giving u p the pawn, and 8 . . . Bf5 9 Nge2 Nc6 10 Ng3 Qd7 1 1 d 4 tries t o hang on t o it, but in n o n e of t hese cases does Black look terribly good to m e .

5 c 3 (D iagram 20) T h e critical move. N o w Black c a n , contrary t o one's first impression, play 5 ... dxc3, since 6 Qa4+ does not win a piece, although Black did apparently resign here in one obsc ure Russian gam e . O n the contrary 6 . . . Nd7 7 Q x e 4 Nc5 followed by . . . Nd3+ i s just fine for Black. 6 bxc3, however, leaves Bl a c k little option but Ba girov's magnificently un ­ abashed 6 ... N c 5 7 d4 Nca6, for example 8 N f 4 e6 9 Nf3 c 5 . Black will continue with s o m ethi ng like . . . cxd4, . . . Bd7, ... Bc6, . . . Nb4 - d 5 . This is very l ittle tested but D a vi e s likes i t for Black once he has come this

1 84

Ot h e r T r i e s f o r W h ite

far. I find i t rather hard to believe Black can really play like th is . In the only game in which it was played White did play 8 Nf4 and subse­ quently exchanged this knight for the a6 kn ight when it arrived on d5, which already looks like a horrible positional error to me, and fol ­ lowed u p with Be4xd5, which looks even worse. White should either relocate h i s e 2 - k n ight to h5 to stop Black dev el o p ing his kingside, or else put it on e 4 . I n de e d 8 Nf4 already looks l i k e t h e wrong square to me. I would have thought 8 Ng3 kept the choice between these two options open longer, as well as making a Qg4/Bg5 h a ck possible. T I P : I n this structure the l a st thing White wants i s to exchange pieces - h e has more space and his chances lie i n a n attack o n the king.

Additionally, Black h a s too many pieces which want to be on d 5 . I n particular once Black has his bishop o n c 6 a n d h i s a6-knight o n d5, the other k night only has the miserable d7-square and, a s Natasha R e ga n pointed out to me after the recent 4NCL game Culverhouse­ Regan, his (or her) queen is sorely pressed for a decent square. On the other hand Black's text move also leaves him under real pressure on a cou ple of fronts.

5 . . . Nc6 This involves a positional pawn sacrifice which I don't really trust.

6 Nxd4 White simply takes the pawn. H i s other dangerous line i s 6 Nf3. Now the natural 6 . . . Bg4 i s u n der a cloud after 7 Nexd4 Nxe5 8 Qe2 Qd5 9 N b 5 ! with the idea of 1 0 c4 . B lack seems to b e bordering on lost here. Instead 6 . . . dxc3 7 bxc3 Bg4 has also been tried, but recently Hector, the knight of this variation for White, demonstrated an effective p l a n with 8 d 4 Q d 5 9 Q c 2 e 6 1 0 Bb2 g 5 1 1 N c 1 ! (simply playing t o rou n d u p t h e e 4 - k n ight) 0-0-0 12 Bd3. T h e n Fritz quite likes Black after 12 . . . Bxf3 1 3 gxf3 Nxc3 1 4 Bxc3 N xd4 1 5 Bxd4 Qxd4 1 6 Rb 1 Qxe5+ 1 7 Kfl, but I think most humans would back the p iece, a n d Hector won shortly.

6 ... Nxd4 6 ... Nxe5? loses to 7 Qe2 Qd5 8 Nb5.

7 Qa4+ c6 8 Qxd4 Qxd4 9 cxd4 N g5 (Diagram 2 1) 10 Bd3

10 Bc4 has been much more common but centralising the bishop on e4 , where it can b e cemented in by d3, seems logical to me. On b 3 i t d o e s nothing e x c e p t w a i t to b e exchanged for the d 5 - b i s h o p , whereas on e4 i t cannot really be e xch a n ged and contributes to White's ki ng­ side play. H ector turned to the text after a rare loss in this line with 10 Bc4 Be6.

1 0 . . . Ne6 White's immediate point was that B lack can't send his bishop to d5 as he would like without first removing h i s knight to c7, since 10 . . . Be6

1 85

S t a rt i n g O u t: A l e k h i n e 's D e fe n c e

1 1 h 4 wins a piece.

Diagram 2 1 Sufficient compensation?

Diagram 22 pawn to

Putting the extra

good use

11 Ne2 g6 12 Be4 Nc7 1 3 b 3 Bh6 14 h4 Be6 1 5 h5 Rd8 Black might consider 15 . . . a 5 and ... a5-a4 here, but i t i s not going to solve hi s basic problem, which i s that his position is cut into two halves and he has less space.

1 6 B b 2 g5 1 7 f3 g4 White has arranged h i s pieces the way h e wants them a n d so has Black, but I don't thin k Black has value for h i s pawn.

1 8 f4 Bd5 19 d3 This is the p o int . White is abl e to avoid the exchange o f light-squared

bishops as bri n ging t h e rear d-pawn to e4 i s unthinkable for Black. When White 's bishop i s on b 3 facing Black's on d5 he menaced by . . . a5-a4, forcing the trade.

is constantly

1 9 ... Ne6 20 g3 Ng7 This looks crazy but is part of a plan which becomes apparent over the next few moves. The knight looked pretty on c7 but in fact it was redundant. Because Black can never contemplate a n excha n g e on e4 t h e knight c a n never reach t h e dream d5-square.

21 Kf2 f5 22 exf6 ep exffi 23 Bxd5 Rxd5 Black has played p urpo sefu lly to force this exchange b u t , in so doing, he has had to put his bishop a n d knight offside, and White is a bl e to change course decisively. 23 . . . cxd5 24 Racl i s worse still .

24 Nc3 Rd8 25 Ne4 Kf7 26 d5 (Diagram 2 2) 26 . . . £5 2 7 B x g 7 Kxg7 27 . . . Bxg7 28 Ng5+.

28 Nc3 cxd5 29 Nb5 a6 30 Nd4 It seems as if Black has won h i s pawn back and may be over the worst, but in fact White enjoys a decisive initiative based on the off.

1 86

O t h e r T r ie s for W h i te

side bishop.

30 ... K f6 31 Rael R e S 32 Rxe8 Rxe8 3 3 Re i B f8 3 4 Re5 Be5 3 5 R x f5 + K e 7 3 6 R x d 5 Bxd 4+ 3 7 R x d 4 Re2+ 3 8 K e 3 R g 2 3 8 . . . Rxa2 3 9 f5 is a l s o hopeless.

3 9 R b 4 Rx g3+ 40 Ke4 b 5 41 a4 b xa4 42 bxa4 Rh3 43 Rb7+ K£6 44 R x h 7 g3 4 5 Ra 7 Rxh5 46 Rxa6+ Kf7 4 7 Kf3 Rd5 4 8 Kxg3 Rx d 3+ 49 Kg4 Rd l 50 Kg5 R g l+ 5 1 Kf5 Rei 5 2 R a 7+ Kg8 5 3 Kg5 R g l+ 54 Kf6 1-0 Hector's realization of his advantage, in keeping with his style, was impressively concrete, a n d it may be that t h i s obscures the fun da· m e ntal features of the position, but I don't think Black can expect to score at all well from the sort of position that arises round about move 16 or 1 7 in this game.

Statistics White has s co red 58% fro m the position after 3 . . . Ne4

Theoretica l ? Not really. Remembering the first six or seven moves an d having an idea of the plans is probably enough, although i t i s worth rem e mber­ ing to steer clear of 5 c 3 Nc6 6 Nf3 Bg4.

Theoreti ca l Statu s Not that good i n m y view. Jonny Hector has been waging war o n 2 . . . d 5 o n a l l fronts, a n d h i s ideas have p u t White well ahead just now. Such Alekhine specialists as Baburin and Varga generally play 2 . . . e5, and I think that's the way to go. I f you don't m i n d the French then 2 . . . d 5 3 e5 Nfd7 (Alekhine's ori gin al method, by the way) is reason­ able, but Black's other methods all look poor to m e .

1 87

I n d ex o f Va r i a t i o n s M a i n Li ne with 4 . . . dxe5 1 e4 N f6 2 e 5 N d 5 3 d 4 d 6 4 N £3 d x e 5 4 . . . c6 40

5 Nxe5 (D iagram) 5 ... c6 (introduction 1 3) 5 . . . g6 (introduction 1 2) 6 Bc4 1 5; 6 c4 24 5 . . . N d 7 (introduction 1 4) 6 Nxf7 44; 6 Nf3 45

6 Be2 30 6 Bd3 32; 6 c4 36

Mai n Li ne w i th 4 . . . Bg4 1 e 4 Nf6 2 e 5 Nd5 3 d 4 d 6 4 Nf3 Bg4 5 Be2 (Diagram) 5 ... e 6 (introduction 49) 5 . . c6 (introduction 5 1 ) .

6 c4 70; 6 Ng5 74 6 0-0 Bxf3 7 Bxf3 dxe5 8 dxe5 e6 9

Qe2 5 1 ; 9 Re 1 68

6 c 4 Nb6 7 Nc3 7 0-0 Be7 8 e x d 6 c x d 6 6 5

7 ... Be7 8 0-0 8 h 3 B h 5 9 exd6 cxd6 10 d5 62

8 0-0 9 h 3 ...

9

B e 3 Nc6 1 0 exd6 cxd6 1 1 d 5 59

9 Bh5 10 Be3 d 5 53 ...

188

I n d ex of Va r i at i o n s

Ma i n Li ne with 4 g6/4 Nc6 . . .

. . .

1 e 4 Nf6 2 e 5 N d 5 3 d 4 d 6 4 N £3 (Diagram) 4 ... g 6 (introduction 78) 4 Nc6 {introduction 79) 5 c4 N b 6 6 e6 9 0 . . .

5 Bc4 N b 6 6 Bb3 Bg7 7 Ng5 80 7 a4 84; 7 Qe2 88

Exchange Vari atio n 1 e 4 Nf6 2 e 5 Nd5 3 d 4 d6 4 c 4 N b 6 5 exd6 (Diagram) 5 ... exd6 (introduction 1 1 5 )

5

. . .

cxd6 (introduction 96)

6 Nc 3 g6

7 B e 3 1 1 0; 7 h3 1 00; 7 a4 1 05; 7 Bd 3 1 0 7; 7 Nf3 97

6 N c 3 B e 7 7 N£3 7 h3 1 1 6; 7 Bd 3 1 2 1 ; 7 Qf3 1 22

7 . 0-0 8 B e 2 Nc6 9 Be3 B g4 ..

115

The Fo u r Paw ns Attack 1 e 4 Nf6 2 e 5 N d 5 3 d 4 d 6 4 c 4 N b 6 5 f4

(introduction 1 26)

(D iagram) 5 . . . dxe5 5 . . . Bf5 1 3 6

5

...

g5 1 46

6 fxe 5 Nc6 6

...

c5 1 41

7 B e 3 Bf5 8 Nc3 e 6 9 N£3 B e 7 9

...

1 26

Bg4 1 31

1 89

S t a r t i n g O u t : A l e k h i n e ' s D e fe n c e

The C hase Vari at i o n 1 e 4 N f6 2 e5 Nd5 3 c 4 Nb6 4 c 5 Nd5 ( D iagram) 5 Nc3 (introduction 1 52) 5 Bc4 1 59

5 ... c6 5

. . .

e6 6 d4 b6 1 59

6 Bc4 d6 1 5 3 6 . . . e 6 1 56

Other Tries for Wh ite 1 e4 Nf6 ( D iagram) 2 e5 2 d3 1 68; 2 Bc4 1 69; 2 Nc3 1 83

2 . Nd5 3 d4 .

.

3 b3 1 68; 3 Nf3 1 68; 3 Bc4 1 68; 3 Qf3 1 68; 3 Nc3 1 72; 3 g3 1 80; 3 c4 Nb6 4 a4 1 67

3 ... d6 4 Bc4 1 65 4 Bg5 1 64; 4 Be2 1 66;

1 90

4

f4 1 69

I ndex of Com p lete Games Ad ams-Agdeste i n , Oslo 1 994 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 76 Adams-Ba b u r i n , Kilkenny 1 997

. . . . . . . .

.

.

.

. . . .

.

.

.

Adams-Burkart, European C l u b Cup 1 996 Almasi.Z-Ke ngis.E, Bern 1 99 6

. .

Berg.E-E ricsso n . J , Skara 2002

. . . . . .

. . . . .

.

.

. .

.

. . . .

Byrne.R-Alburt, Berkeley 1 9 84

.

.

. .

. .

.

. . . .

. . . .

B o l o gan-Agd e s t e i n.S, Yerevan 1 996 Borriss-B a b u r i n , Berlin 1 9 9 2

.

. .

.

. . .

. .

. . . . .

.

. .

.

..

.

..

.

.

. . .

. . . .

. .

.

. .

. . .

.

. . .

. . .

. .

.

. .

.

. . .

.

.

. . .

.

. .

.

. . . .. .

.

. .

.

.

. . .

.

.

.

.

. . .

.

. .

D i mitrov.V-M arinkovic, Tivat 1 9 9 5 Emms-D a v i e s , Southend 2 00 2 Ferguson-Ne i1, Norwich 1 9 9 4

. .

. .

.

.

.

.

.

. .

.

. . .

.

..

. .

.

.

. .

.

.

.

G a l l a gh e r- C ro u ch, 4NCL 2 002 .

. .

.

.

.

. .

G d anski-Marinkovic, Biel 1 989

. .

.

.

.

.

. . . . .

Gomes-S o p p e , I tau Cup 2 0 0 1

.

.

. .

. . . . .

. .

. .

. .

. .

Grischuk-Bab u r i n , Torshavn 2000

. .

. .

. .

. .

.

.

.

. . .

. .

.

.

.

.

. .

. .. . . .

..

.

. . . .

.

.

. . .

.

.

. . .

. . .

Hector-An a g n o s t o u , Komotini 1 992

.

.

. .

.

.

.

. .

...

.

. .

.

. . . .

.

. .

. . .

. . . . . . .

G r ischuk-Ponomariov, Torshavn 2000 H a l s e r-Ko n o p k a, Graz 2 0 0 1

. .

.

.

.

. .

. .

.

.

. .

.

. . .

. .

. .

.

.

.

.

. .

.

. . . .

.

. .

. . ..

.

.. .. .

.

. . . .. . . . . .

. . .

. . . .

.

.

.

.

.

.

.

.

.

.

. .

.

. . . . .

. . . .

. .

. . . .

.

.

. .

.

.

.

.

.

. .

.

.

. .

. .

.

. .

.

. . .

.

.

. . .. . .

. .

.

. .

. . . . . . .

.

.

.

.

. .

.

.

.

. .

.

.

. .

.

.

.

.

.

. .

. . . .

.

.

.

.

.

. . .

. . .

. .

.

.

.

. .

.

.

.

. . . .

. .

. . .

. . .

. .

.

.

.

.

.

. .

. .

.

. . . . .

.

.

. .

. .

. . .

. .

. . . . . . . .

.

. .

.

.

.

. . .. . .

.

.

. .

. .

.

.

. .

.

. .

. .

.

. .

.

. . . .

.

.

. . .

.

.

. .

. .

.

.

. . .

.

.

.

. . .

.

. . .

. .

.

.

. . .

.

. .

. .

.

.

. .

.

.

. . .

. .

. .

.

.

.

.

. . .

.

. .

. . . . . .

. . . . .

.

.

.

.

. .

.

. .

. .

.

.

. . . . .

.

. . . . . . . . .

.

. . .

.

. . . . .

. . .

.

.

.

.

.

.

.

.

.

.

. .

.

. .

. .

.

. . .

. . .

.

.

. .

.

. .

. .

. . .

. .

.

. .

.

.

.

.

. .. .

. .

. .

.

. .

. .

. .

. .

. . .

. . . .

. .

. . . . . .

. .

. .

.

. .

.

. . . .

.

. .. .

. . . . . . . .

. . .

. . . . . . .

.

..

.

.

.. .

.

.

. .. . .

. .

.

. . . . .

. .

. . .

.

.

..

..

.

. .

1 22

.

. . . . . .

..

.

.

.

.

.

. .

62

. . . . . . . . . .

.

..

59

.

.

. .

. . .

. .

. .

.

.

. . . . .

.

.

.

.

.

. . .

.

. .

. .

. . .

. .

. .

.

..

.

.

. . .

. .

.

. . . .

.

.

.

.

. . . . . . . .

. . .

. .

.

.

.

.

.

. .

.

.

.

. . . .

.

.

. . .

. .

.

. . .

.

131 .

.

.

. .

. . . .

. . . . . .

. .

.

. .

.

. . . . . . . . . .

.

. . . . . . .

.

. .

.

. . .

.

. .

. . . .

.

.

. .

.

. .

.

. . .

.

. .

..

. . .

.

.

. .

. .

. 88 .

1 20

. . . . .

. .

.

. ..

.

...

. .

.

.

.

.

.

. .

.

. .

80

. .

115

.

. .

. . . . .

.

. .

96

..

24

.

84

. . . . .

. .

. . .

.

1 56

. . .

1 83

.

127

. . . . .

.

. .

. .

. . .

.

...

. .

40

. . . . . . .

.

.

. .

70 68

.

.

.

.

. .

28

. . .

.

.

.

.

. .

.

.

.

. .

. . .

. .

.

. .

.

. .

.

.

. . .

. . . . .

Hoskyn-D ive, New Zealand Correspondence C h a m pionship 1 997 Kall a i-Hor vath . C , Bu dapest 1 990

..

15

110

. . .. .

. . . .

.

. . . . .

. . .

. . . . . . . . .

. . .

. .

..

. . . . . .

19

.

. .

.

36

.

. . .

. .

.

.

118

.

.

. . . . .

. .

..

.

. . . .

.

.

.

. .

. .

. .

. .

. . .

.

. .

.

.

. . . . . . . .

.

.

.

.

.

.

. .

. . . .

. . .

.

. . . . . . . . . . . . . . . . .

. . . . .

.

. .

.

. . . . . . . . . . . . . .

. . . .

.

.

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .

G a ll a g h e r-Babu r i n , Isle of M a n 2 0 0 1

G i p s l i s-Larsen, Sousse 1 967

.

.

. . .

C o l eman.D-Hi l l arp-Persson, Copenh agen 1999 David .A-Bauer. C , C l i c h y 2004

.

.

. . .

. . .

..

.

.

.

. .

.

.

. . . . . . . . . . . . . . . . . . . . . . . .

C h r i s t i a n s e n.L-Ke n g i s . E , M a n i l a 1 992 .

. .

. . .

. .

.

.

.

.

.

.

.

. .

. .

. .

. . .

. 43 .

191

S t a r t i n g O u t : A l e k h i ne ' s D e fe n c e

Khmelnitsky- P a l atnik, Kherson 1 989 . . . . . . . . . . . . . . .

Kra n z l-Baumgardt, I C C F e - m a i l 2000

.

.

..

.

.

.. . .

. .

.

. .

. .

. .

....

. .

. . . .

. .

.

. .

.

.

.. ...

. .

.

Kupre i c h i k-Var ga . Z , European C h ampionship 1 992 . . . . . Maj ostorovic-Bry s o n , Correspondence 1 98 5

.

M o v s e s s i a n .S-Lut her, Istanbul 2003 . . . . . . . . . . . .

.

. . . . .

.

. . .

. . ...

.. .

. . . ..

.

..

. .. .

.

..

. . . . . .

.

. . . . 1 00 .

..

. .

.

. . . . 1 46

.

1 69

........ ....... . . . . . .

.

1 36

. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .

. .

. .. ..

.

.

........

.

. . . . .

..

.

.

.

1 41

......... ......

Pade vsky-V u k o v i c. M , Smederevska Pala n ka 1 9 7 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 72 P i za-M i ro s h n i c h e n k o , Komercni Banka O p e n 1 996 . . . . . . . . . . . . . . . . . . . . . . . . . . 4 5 .

.

.

.

. .

.

Rogovski-Murashko, Alushta 200 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

Rozental i s-Adams, Bun desliga 1 99 8 . . . . . .

. .

.

.

.

...

. .

.. .

.

.

.

. .. . . .. ... .. . .. . . ......

1 59

. .

..

.

. .

.

. . . . .

.....................................

1 05

S a n akoev-Davi d . G , Correspondence 1 98 8 - 9 2 . . . . . . . .

.

.

. . .

. .

. . . .

... .. ... . .

.

. . .

....

.

74

1 80

........

.

. .

. ..... .. ..

.. ........ . .. .. ..

Rozental i s-Kuczyn s k i , B a n k Pocztowy Open 200 1 Salmens u u-Solo z h e n k i n , Fin l a n d 2000 .

.

.

. . .. . .

. . .

.

.

. .

. 53

.. .

.

Skasl i e n-C aranzano, Correspondence 1 988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 .

S p r a ggett. K-Nataf, Enghi e n - l e s - Ba i n s 1 99 7 . . . S utovsky- C a rls e n . M , Rethy m n o n 2003 . . . . . . . . Thorh a l l son-Bu r ge s s . G , Gausdal 1 9 9 7 Z u i d ema-H a a k e rt, Den H a a g 1 9 85

1 92

. .

...

. .

.

. .

.

... .. .

. .

.

.

... . .....

.

..

. . .

.

... .. . .... .

..

.

... .. . .. ... .

. .

. . .

....

. . ........... ...

.............. .

.

.

...........

. .

. .

. .

. .

...

...

..

.

107

. . . .

...

32

. 1 52

. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .

90